Iron therapy supplementation in inflammatory bowel disease patients with iron deficiency anemia: findings from a real-world analysis in Italy

Abstract

Background

This real-world analysis evaluated iron therapy supplementation in inflammatory bowel disease patients with iron-deficiency anemia, considering disease progression and healthcare resource consumption.

Methods

A retrospective observational study was conducted using administrative databases of a pool of Italian healthcare entities, covering about 9.3 million beneficiaries. Between January 2010 and September 2017, adult patients were enrolled in the presence of either hospitalization or active exemption code for ulcerative colitis/Crohn’s disease, or one vedolizumab prescription. Iron-deficiency anemia was identified by at least one prescription for iron and/or hospitalization for iron-deficiency anemia and/or blood transfusion (proxy of diagnosis). Patients were divided in untreated and iron-treated during 12-month follow-up and analyzed before and after propensity score matching. Disease progression, was evaluated through inflammatory bowel disease-related hospitalizations and surgeries, and healthcare resource utilization was assessed.

Results

Overall, 1753 patients were included, 1077 (61.4%) treated with iron therapy and 676 (38.6%) untreated. After propensity score matching, 655 patients were included in each group. In unbalanced cohorts, disease progression was significantly reduced in patients receiving iron therapy compared to the untreated (11.0% vs. 15.7%, P < 0.01), and this trend was maintained also after applying propensity score matching. The overall mean cost/patient was significantly lower in iron-treated than untreated (4643€ vs. 6391€, P < 0.01).

Conclusion

The findings of this real-world analysis suggest that iron therapy was associated with significant benefits in inflammatory bowel disease patients with iron-deficiency anemia, in terms of both disease progression and healthcare resource utilization.

Introduction

Crohn’s disease (CD) and ulcerative colitis (UC) are inflammatory disorders of the gastrointestinal tract, belonging to the overarching inflammatory bowel diseases (IBD) [1].

IBD mainly affects young people, of both genders, with comparable incidence (age range 15–40 years) [2,3]. Being a chronic disease, IBD generates a significant reduction in the quality of life [4] and high consumption of healthcare resources (medications, outpatient visits, hospitalizations, and surgeries) [5]. IBD is characterized by chronic inflammation of the bowel and by several extra-intestinal manifestations which are present in 25–40% of IBD patients [6]. Anemia is a high incidence manifestation of IBD [7] and about one-third of patients with IBD present hemoglobin levels below the threshold of 12 g/dl for females and 13 g/dl in males [7]. In IBD patients, the most common cause of anemia is iron deficiency (called iron deficiency anemia, IDA) [8]. The pathogenetic explanation of IDA in IBD patients could be related to several factors, which result from chronic blood loss and/or reduced iron absorption and reduced iron intake [9]. IDA subsequent to IBD is associated with reduced quality of life, an increase in work absenteeism, an increased frequency of hospitalization, and increased healthcare costs [4,5]. Thus, there is a clear need to support IDA patients with pharmacological treatments, to improve their quality of life and physical condition, and alleviate fatigue and cognitive deficits [10]. Pharmacological intervention options include oral or intravenous iron therapy. However, the efficacy of several oral iron therapies might be limited in gastrointestinal inflammation due to the impairment of iron absorption by enterocytes and might not be able to compensate for the ongoing blood loss [11]. Additionally, several oral iron preparations like Fe-fumarate or FE-sulphate could often be poorly tolerated [12,13]. Thus, intravenous iron administration is preferred in IBD patients [14,15]. The combination of iron deficiency and chronic inflammation may result in significant anemia, which must be considered during patient management and therapy. Current European Crohn’s and Colitis Organization guidelines recommend the identification and treatment of anemia in IBD [15], with treatment associated with improved quality of life [16,17]. Because anemia is associated with increased disease severity and healthcare resource utilization and costs [18], treatment of anemia with iron supplementation could be expected to lower both the clinical and economic IBD burden. But currently, very limited data on the effect of iron therapy in IBD disease progression and healthcare resource consumption have been collected from real-world clinical practice. Thus, the present retrospective analysis aimed to evaluate the impact of iron therapy on disease progression in IBD patients with IDA and healthcare resource consumption and costs, in Italian settings of clinical practice.

Methods

Data source

In this retrospective observational comparative study, data were collected from Italian Healthcare Departments’ administrative databases, covering around 9,3 million health-assisted individuals. The following databases were used: (1) demographic database, which contains all patient demographic data, such as gender, age and date of death; (2) pharmaceuticals database, that supplies information on medicinal products reimbursed by the National Health System (NHS), as the Anatomical Therapeutic Chemical (ATC) code, number of packages, number of units per package, unit cost per package, and prescription date; (3) hospitalization database, which encloses all hospitalizations data for patients under analysis, such as the discharge diagnosis codes classified according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), Diagnosis Related Group (DRG) and DRG related charge (provided by the Italian Health System); 4) outpatient specialist services database, which incorporates all information about visits and diagnostic tests for patients under analysis (date and type of prescription, description activity and laboratory test or specialist visit charge); (5) payment exemption database, which contains data of the exemption codes that allow to avoid the contribution charge for services/treatments when specific diseases are diagnosed. An anonymous univocal numeric code was assigned to each study subject to guarantee patients’ privacy, in full conformity with the European General Data Protection Regulation (2016/679). The patient code in each database permitted the electronic linkage among all databases. All the results coming out from the analyses were produced as aggregated summaries, which cannot assign, either directly or indirectly, to individual patients. The study was conducted according to the principles of the Declaration of Helsinki the study and its updates, and approved by the following Institutional Review Board (or Ethics Committees): (1) Comitato Etico Indipendente Azienda Ospedaliero-Universitaria ‘Consorziale Policlinico’ (Bari) (Prot. N 0059036/09/07/2019, 09/07/2019); (2) Comitato Etico Sezione Area Centro – Regione Calabria (Prot. N 43, 22/02/2019); (3) Comitato Etico Lazio 2 (Prot. N 0087354, 15/05/2019; Prot. N 0047061/2020, 12/03/2020); iv) Comitato Etico per la Sperimentazione Clinica delle Province di Verona e Rovigo (Prot. N 50045, 25/07/2018). Informed consent was not required since obtaining it was impossible for organizational reasons (pronouncement of the Data Privacy Guarantor Authority, General Authorization for personal data treatment for scientific research purposes – n.9/2014).

Study design and study population

All adult (≥18 years old) patients diagnosed with IBD were included in this study between 01/01/2010-30/09/2017 (inclusion period). The inclusion criteria were: (1) the presence of UC (identified by at least one hospitalization with a primary or secondary diagnosis with ICD-9-CM code 556 and/or active exemption code 009.556) and/or (2) CD (identified by at least one hospitalization with a primary or secondary diagnosis with ICD-9-CM code 555 and/or active exemption code 009.555) and/or (3) that have received at least one prescription for Vedolizumab (ATC code L04AA33). The IBD diagnosis date was defined as the date of first hospitalization, drug prescription, or active exemption for IBD during the enrollment period. Twelve months before the IBD diagnosis date was used as a patient characterization period. Among IBD patients, those with IDA diagnosis (IBD-IDA patients) were identified by the presence of at least one prescription for iron preparations (ATC code B03A) and/or at least one hospitalization for IDA (primary or secondary diagnosis with ICD-9-CM code 280) and/or at least one blood transfusion (ICD-9-CM code 99.0, primary or secondary procedure, ICD-9-CM code V58.2, primary or secondary diagnosis, specialistic code 99.07.1) during the 12-month-period after IBD diagnosis date. For all IBD-IDA patients, the index date started 12 months after IBD diagnosis date and marked the beginning of 1 year follow-up period. IBD-IDA patients were divided into two subgroups: (1) patients with at least one prescription of iron preparations during a 12-month-period after the IDA inclusion date, and (2) patients without any prescription of iron preparations during 12 month-period after IDA inclusion date. Patients who die before the index date or during the follow-up period were excluded.

Baseline patients’ characteristics

At IBD diagnosis, age, gender, and disease type [CD (patients identified by CD inclusion criteria) or UC (patients identified by UC inclusion criteria) or undefined (patients identified by vedolizumab use as inclusion criteria)] has been identified. During the characterization period, the following diagnoses/comorbidities were assessed: cardiovascular diseases (identified by hospitalization discharge diagnosis code ICD-9-CM codes 410, 411, 413, 414, 430-438, 440, 443); chronic obstructive pulmonary disease (COPD) (identified by hospitalization discharge diagnosis code ICD-9-CM codes 490-492, 494, 495, 496, or a prescription of medication with ATC code R03 in patients aged >40); asthma (identified by hospitalization discharge diagnosis code ICD-9-CM code 493 or a prescription of medication with ATC code R03 in patients aged ≤40); cancer (identified by hospitalization discharge diagnosis code ICD-9-CM codes 140-239); infections (identified by a prescription of drug with ATC code J01); diabetes (identified by a prescription of drug with ATC code A10); hypertension (identified by a prescription of drug with ATC codes C02, C03, C07, C08, C09); kidney disease (identified by hospitalization discharge diagnosis code ICD-9-CM codes 585.2, 585.3, 585.4); rheumatoid arthritis (identified by hospitalization discharge diagnosis code ICD-9-CM code 714.0 or active exemption code 006.714.0); ankylosing spondylitis (identified by hospitalization discharge diagnosis code ICD-9-CM code 720.0 or active exemption code 054.720.0); psoriasis (including psoriatic arthritis) (identified by hospitalization discharge diagnosis ICD-9-CM codes 696.0, 696.1, 696.2 or active exemption codes 045.696.0, 045.696.1); menorrhagia, metrorrhagia (identified by hospitalization discharge diagnosis ICD-9-CM codes 626.6; 627.0);}; disease severity (using score from 0 to 4, described in Table 1); number of similar treatments at inclusion [searched as the presence of at least one prescription, during the 12 months before the index date, of biologicals disease-modifying antirheumatic drugs, antidiarrheals (details listed below), and all antibacterial for systemic use [ATC code J01)]. Patients have also been characterized based on the Charlson comorbidity index, which assigns a score to each concomitant disease (other than IBD) assessed during the characterization period [19].

Table 1.

Criteria for the definition of IBD severity

Description	Severity grade
- No previous IBD hospitalizations, no previous IBD treatment	0
- No previous IBD hospitalizations, previous treatment with only mesalazine or sulfasalazine or budesonide	1
- No previous IBD hospitalizations, previous IBD treatment (except mesalazine, sulfasalazine, budesonide)	2
- Previous IBD hospitalizations, no previous IBD treatment	3
- Previous IBD hospitalizations, previous IBD treatment	4

IBD progression evaluation

During the follow-up period, the IBD progression was evaluated and defined as the presence of at least one of the following proxies: (1) IBD-related hospitalizations (CD or UC, ICD-9-CM code 556 or 555); or (2) IBD-related surgical interventions (proctocolectomy surgery, ICD-9-CM procedural codes 45.3X, 45.4X, 45.6X, 45.7X, 45.8X, 45.9X, 46.0X, 46.1X, 46.2X, 46.3X, 46.6X, 46.7X, 48.0X, 48.1X, 48.3X, 48.4X, 48.5X, 48.6X, 48.7X, 48.8X, 48.9X, 49.0X, 49.1X, 49.3X, 49.5X, 49.6X, 49.7X).

Statistical analysis

Continuous variables were reported as mean ± SD, categorical variables were expressed as numbers and percentages. IBD-IDA patients were categorized in the two cohorts, patients treated with or without iron preparation. The results were compared between the two cohorts by using Student’s t-test (to compare continuous variables) and chi-square test (to compare categorical variables) and statistical significance was accepted at a P-value <0.05. Given the study’s retrospective observational nature, the non-random assignment of patient among different cohorts can lead to not comparable groups on several characteristics based on existing data. Thus, the propensity score matching (PSM) methodology was applied to abate potential unbalances in baseline characteristics among the two cohorts (IBD-IDA patients treated with or without iron therapy). The propensity score was estimated using a logistic regression model, considering all the baseline confounding variables (listed above) and patients were matched on quintiles of propensity score. To maintain the maximum number of patients, a 1 : 1 matching algorithm was used; that is, for one IBD-IDA patient without iron therapy, 1 IBD-IDA patient with iron therapy was sampled. To examine the balance of covariate distribution among the two cohorts after PSM, P-value and standardized mean difference (SMD) were evaluated. SMD greater than 0.1 (10%) is a threshold being recommended for declaring imbalance [20]. All analyses were performed using Stata SE version 17.0 (StataCorp, College Station, Texas, USA).

Results

From the original 14,349 IBD patients identified, 13 475 were selected following the inclusion and exclusion criteria of the study. Among them, patients with a diagnosis of IDA (IBD-IDA) resulted to be 1,753; the cohort were subdivided into two groups: i) IBD-IDA patients treated with iron therapy (n = 1077, representing 61.4% of the total number of IBD-IDA patients) and ii) IBD-IDA patients without iron therapy (n = 676, representing the 38.6% of overall IBD-IDA patients). The baseline characteristics of IBD-IDA patients treated or not with iron therapy (both oral and parenteral preparations), before and after PSM, are reported in Table 2.

Table 2.

Demographic and clinical characteristics of IBD-IDA patients treated or not with iron therapy, before and after PSM

	Before PSM			After PSM
	Without iron therapy	With iron therapy	P-value	Without iron therapy	With iron therapy	P-value	SMD
Patients, n	676	1077		655	655
Age at IBD diagnosis, years	58.5 ± 20.1	54.5 ± 20.2	<0.001	58.3 ± 20.0	58.0 ± 20.4	0.822	0.012
Male gender	293 (43.3%)	481 (44.7%)	0.589	289 (44.1%)	304 (46.4%)	0.405	0.046
Charlson comorbidity index	1.1 ± 1.4	1.0 ± 1.3	0.354	1.1 ± 1.4	1.1 ± 1.4	0.888	0.008
Cardiovascular disease	62 (9.2%)	96 (8.9%)	0.854	62 (9.5%)	67 (10.2%)	0.643	0.026
COPD	94 (13.9%)	170 (15.8%)	0.377	100 (15.3%)	101 (15.4%)	0.939	0.004
Asthma	24 (3.6%)	45 (4.2%)	0.510	22 (3.4%)	24 (3.7%)	0.764	0.017
Diabetes	87 (12.9%)	138 (12.8%)	0.973	85 (13.0%)	82 (12.5	0.804	0.014
Hypertension	343 (50.7%)	469 (43.5%)	<0.01	334 (51.0%)	330 (50.4%)	0.825	0.012
Cancer	73 (10.8%)	87 (8.1%)	0.054	70 (10.7%)	71 (10.8%)	0.929	0.005
Infections	402 (59.5%)	643 (59.7%)	0.922	391 (59.7%)	388 (59.2%)	0.866	0.009
Kidney disease	18 (2.7%)	20 (1.9%)	0.260	16 (2.4%)	17 (2.6%)	0.860	0.010
Rheumatoid arthritis	9 (1.3%)	14 (1.3%)	0.955	9 (1.4%)	9 (1.4%)	1.000	0.000
Ankylosing spondylitis	9 (1.3%)	6 (0.6%)	0.087	7 (1.1%)	4 (0.6%)	0.364	0.050
Psoriasis/psoriatic arthritis	7 (1.0%)	12 (1.1%)	0.877	7 (1.1%)	6 (0.9%)	0.780	0.015
Previous IBD intervention	120 (17.8%)	142 (13.2%)	0.428	116 (17.7%)	112 (17.1%)	0.771	0.035
IBD severity score	3.7 ± 0.6	3.1 ± 1.2	<0.01	3.7 ± 0.6	3.6 ± 0.6	0.529	0.012
Type of IBD diagnosis						0.920	0.023
Ulcerative colitis	432 (63.9%)	687 (63.8%)	0.415	418 (63.8%)	420 (64.1%)
Crohn’s disease	241 (35.7%)	379 (35.2%)		234 (35.7%)	231 (35.3%)
Undefined	NI	11 (1.0%)		NI	4 (0.6%)
Hepatic cirrhosis	19 (2.8%)	26 (2.4%)	0.609	19 (2.9%)	16 (2.4%)	0.607	0.028
Number of IBD drugs	1.8 ± 1.2	1.9 ± 1.2	<0.01	1.8 ± 1.2	1.8 ± 1.2	0.962	0.003

Continuous variables were reported as mean ± SD, categorical variables as numbers and percentages. Significant P-values are highlighted in bold.

Before PSM analysis, the two cohorts, composed of 676 patients without iron therapy and 1077 with iron therapy, were comparable for almost all variables, except, that is, age, incidence of hypertension, previous IBD interventions, and IBD drugs prescribed. The mean age for IBD-IDA patients without and with iron therapy was 58.5 ± 20.1 and 54.5 ± 20.2, respectively (P < 0.001), and the 43.3% and 44.7%, respectively, were male. Most IBD-IDA patients, treated or not with iron therapy, had hypertension (50.7% and 43.5%, respectively, P < 0.01), and infections (59.5% and 59.7%, respectively). In comparison, only a minority were characterized by previously COPD (14.9% and 16.5%, respectively), diabetes (12.9% and 12.8%, respectively), cardiovascular events (9.2% and 8.9%, respectively), cancer (10.8% and 8.1%, respectively). After PSM with 1:1 balancing, both cohorts of IBD patients without and with IDA therapy consisted of 655 subjects, and resulted in statistically comparable, indicating that the two groups were well-balanced.

During the follow-up period, the IBD progression was evaluated and defined as the presence of at least one of the following proxies: (1) IBD-related hospitalizations; or (2) IBD-related surgical interventions. As reported in Table 3, the incidence of IBD-related hospitalization was significantly lower in patients receiving iron therapy (10.5% vs. 14.3%, P < 0.05); also, in these patients, a slightly, but not statistically significant decrease of IBD-related surgical interventions was found; the overall IBD progression, was significantly lower in patients under iron therapy vs. the untreated cohort (11.0% vs. 15.7%, P < 0.01). After applying PSM methodology, in IBD-IDA patients treated with iron therapy, the frequency of IBD-related hospitalization, IBD-related surgery and the overall IBD progression tended to be lower than that in the untreated group, but without statistical significance.

Table 3.

IBD progression during follow-up period in IBD-IDA patients treated or not with iron therapy, before and after propensity score matching

	Before PSM			After PSM
	Without iron therapy	With iron therapy	P-value	Without iron therapy	With iron therapy	P-value
Patients, n	676	1077		655	655
IBD-related surgical interventions	27 (4.0%)	26 (2.4%)	N.S.	26 (4.0%)	21 (3.2%)	0.458
IBD-related hospitalizations	97 (14.3%)	113 (10.5%)	<0.05	97 (14.8%)	80 (12.2%)	0.169
IBD progression	106 (15.7%)	119 (11.0%)	<0.01	105 (16.0%)	86 (13.1%)	0.137

Variables were expressed as numbers and percentages.

A multivariate regression analysis was performed to estimate predictors, among baseline variables, of IBD disease progression. As shown in Table 4, the older age at IBD diagnosis (OR 0.984, 95%CI: 0.974–0.995, P < 0.01) was slightly negatively associated with IBD progression, while the presence of ankylosing spondylitis (OR 4.175, 95%CI: 1.121–15.545, P < 0.05) and the number of IBD medications (OR 1.556, 95%CI: 1.283–1.887, P < 0.001) were positive predictors of IBD progression in IBD-IDA patients during the follow-up period. Also, the prescription of iron therapy tended to be negatively associated with the IBD progression (OR 0.792, 95%CI: 0.576–1.091, P = 0.153), although it did not reach the statistical significance.

Table 4.

Logistic regression model to identify predictors of IBD progression in patients with IDA

Covariates	OR	95% conf. interval		P-value
Age at IBD diagnosis	0.984	0.974	0.995	<0.01
Gender (ref. female)	1.267	0.917	1.752	0.152
Charlson Index	1.008	0.818	1.244	0.938
Cardiovascular disease (ref. absence)	0.602	0.294	1.234	0.166
COPD (ref. absence)	0.608	0.343	1.076	0.088
Asthma (ref. absence)	0.912	0.401	2.073	0.826
Diabetes (ref. absence)	0.795	0.409	1.546	0.500
Hypertension (ref. absence)	1.277	0.835	1.952	0.260
Cancer (ref. absence)	0.762	0.343	1.692	0.504
Infections (ref. absence)	0.723	0.483	1.082	0.115
Renal insufficiency (ref. absence)	1.289	0.461	3.604	0.629
Rheumatoid arthritis (ref. absence)	0.498	0.101	2.444	0.390
Ankylosing spondylitis (ref. absence)	4.175	1.121	15.545	<0.05
Psoriasis/psoriatic arthritis (ref. absence)	1.560	0.430	5.664	0.499
Previous IBD intervention (ref. absence)	0.882	0.577	1.349	0.563
IBD severity	1.196	0.837	1.708	0.325
Type of IBD diagnosis
Ulcerative colitis (ref.)	1.000
Crohn’s disease	1.543	1.100	2.164	<0.05
Undefined	0.697	0.076	6.382	0.749
Hepatic cirrhosis (ref. absence)	0.784	0.260	2.365	0.665
Number of IBD drugs	1.556	1.283	1.887	<0.001
Iron therapy
No	1.000
Yes	0.792	0.576	1.091	0.153

Significant P-values are highlighted in bold.

The overall healthcare resource consumption by IBD-IDA patients, including drugs, hospitalizations, and outpatients’ services, is shown in Fig. 1. In IBD-IDA patients receiving iron therapy, a significant reduction of the mean annual number of hospitalizations (0.5 vs. 0.6, P < 0.01) and outpatient specialist services prescriptions (8.7 vs. 11.4, P < 0.001) was found, respect to untreated patients (Fig. 1a). After applying the PSM methodology, the mean annual number of outpatient specialist services prescriptions (9.2 vs. 11.5) was still significantly lower in patients treated with iron therapy vs. the counterpart cohort (Fig. 1b).

Fig. 1.

Mean annual healthcare resource consumption during follow-up period in patients with IDA: treated and not treated with iron therapy. (a) Pre-PSM; (b) post-PSM.

The annual mean of drug prescriptions resulted in being moderately, but statistically significant, higher in IBD-IDA patients treated with iron therapy (17.2 vs. 15.8, P < 0.01), only after applying PSM (Fig. 1b). Nevertheless, this result was not associated with a statistically significant increase in drug-related costs for IDA treated patients, both before and after PSM analysis (Fig. 2). In iron-treated and untreated patients respectively, before applying PSM methodology, the costs related to outpatient services (773€ vs. 1.216€, P < 0.05) and the total healthcare costs (4.643€ vs. 6.391€, P < 0.01), during the follow-up period, were significantly lower (Fig. 2a). The same trend was observed after PSM analysis, but without reaching statistical significance (Fig. 2b).

Fig. 2.

Mean annual costs during follow-up period in patients with IDA: treated and not treated with iron therapy (a) pre-PSM; (b) post-PSM.

Discussion

This retrospective analysis of administrative claims data gave insights into the impact of iron supplementation on disease progression and healthcare resource consumption in anemic IBD patients, in a real-world clinical Italian practice setting. Among the study population of anemic IBD patients, almost 60% of patients were under iron supplementation therapy, while 40% did not receive any iron-based drug prescription during the one-year follow-up period after IDA diagnosis. The literature has already reported the undertreatment of IDA IBD patients. In particular, Goodhand et al. [2121] showed that only a tiny proportion of anemic IBD patients received oral (children 13%, adolescents 30%, adults 48%) or parenteral iron supplementation (adolescents 30%, adults 41%) [21]. According to the European Crohn’s and Colitis Organization (ECCO) recommendations, iron supplementation is recommended for all patients with anemia associated with iron disorder [15]. In this analysis, we also included blood transfusions among the criteria as a proxy to identify IDA. The ECCO guidelines suggest restricting this procedure to some acute condition like important blood loss or surgery to rapidly achieve an increase of Hb or to restore a hypovolemic situation [15], and indeed there is large body of evidence to indicate that transfusion requirements wiped out after the advent of IV iron and ESA [22]. In this analysis, anyhow, transfusions were included in view of the long observational time, to avoid missing those patients that might have been received a transfusion in the past, when there was a more flexible attitude towards transfusion administration in anemic IBD patients.

Although these recommendations, current observational data suggest that it remains underdiagnosed and undertreated [23,24]. Thus, leaving IDA undertreated could worsen patient clinical condition, since it has been reported that IBD patients with persistent or recurrent anemia had higher indices of disease activity, exhibit lower average quality of life and required significantly more healthcare resources, than patients without anemia [25].

The analysis of demographic and clinical patient characteristics at baseline has shown that the two cohorts, without and with iron therapy, were comparable and differed for the age at study inclusion (54.2 and 58.5 years in patients with and without iron therapy, respectively), and, that is, for the lower incidence of hypertension, previous IBD surgical interventions, and other variables. Due to the non-randomly allocation of patients among the iron-treated and untreated cohorts, PSM methodology was applied to the baseline covariates of the two study groups to balance treated and untreated patients [20]. After applying PSM, completely balanced cohorts were defined. Among overall patients, the treatment with iron therapy was associated with a significant reduction of IBD disease progression, especially in terms of IBD-related hospitalization; in the covariate-balanced cohorts, this trend was still observed, although the statistical significance was not reached. In addition, a trend of 20% risk reduction of IBD progression in patients under iron therapy was observed. Much evidence in the literature showed that, in real-world settings, the recurrence of anemia in IBD patients is correlated with repeated hospitalizations, medications, and IBD-related surgeries [26,27]; in addition, comorbid anemia in IBD individuals is correlated with inflammatory activity and may occasionally both precede the development of intestinal symptoms and be the key signal unmask IBD [27]. Our data suggest that IDA treatment might be beneficial for IBD patients in terms of disease progression. In IBD patients, anemia is also associated with increased healthcare utilization and costs, thus, treatment of anemia with iron supplementation could be expected to restrain the economic disease burden. In the present study, the analysis of healthcare economic impact of iron treatment in IBD patients’ management showed a significant reduction of outpatient specialistic service prescriptions (in terms of specialistic visits, and diagnostic tests) in treated vs. untreated patients. The total mean direct costs were significantly lower in patients under iron therapy, but only a slight difference in mean annual cost values was found among covariate-balanced cohorts. Similar to other reports [28], in this analysis, the treatment of anemia was not associated with a significant reduction of direct healthcare costs. A possible explanation could be related to the fact that in this clinical setting, the treatment with iron supplementation could be associated with some beneficial effect, without directly impacting healthcare costs: as reported by Rieder et al. [29], in complicated CD patients, the beneficial effect of iron therapy was observed in the reduction of disease complications or need for surgery [29]. The present analysis was not designed with the objective to discriminate among different iron formulations; it is well recognized that the use of oral vs. intravenous preparations could have a different impact on disease progression and cost-benefit balance in IBD patients, which directly depend on the rapidity of anemia improvement by treatment, its implications for comorbidities, hospitalization rates and patient quality of life improvement [30]. Even if under iron supplementation, patients could still have high iron deficit levels that may request more outpatient visits and consumption of healthcare resources.

The present results should be interpreted by considering some limitations related to the retrospective nature of the current analysis. Administrative databases have progressively improved the quality of the collected data. However, some information may be missing, and patients with missing data were excluded from the analysis. We were not able to include clinical data on disease severity (by test laboratory assessments). Moreover, IBD progression was ascertained using IBD-related hospitalizations and surgeries. In spite of the advantages of the use of administrative databases related to the large numbers of available data, as mentioned before, this approach might suffer from some unavoidable flaws, since some diagnoses or conditions can be traced only through indirect tools (like hospitalization discharge codes, procedure codes or ATC codes for drug therapies) that are indeed a proxy of diagnosis. In addition, data on disease location and behavior (especially for CD) were not retrievable. Moreover, besides hospitalization and exemption codes for UC and CD, we used the presence of vedolizumab therapy among inclusion criteria to identify all IBD patients, also those without an uncertain diagnosis: the reason beyond this choice is the reimbursement by the Italian National System during the period of inclusion of the patient population, from January 2010 to September 2017. Vedolizumab has been first indicated (2014) and then approved for reimbursement (2016) by the AIFA for the treatment of IBD [31], and in addition, it might be considered vedolizumab as the best tracer to detect IBD since other treatments, either conventional or biological, are approved also for the treatment of other immune-mediated inflammatory conditions, like rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, psoriasis, juvenile idiopathic arthritis, uveitis and hidradenitis suppurativa [32]. This might have led to the exclusion of some eligible patients. In the PSM analysis, the presence of some confounding factors that can affect disease progression (i.e. smoking habit) were not included since they were not available in the database, representing a study limitation. Comorbidities and other diagnoses (including IDA) were evaluated by using a proxy of diagnosis; there might be an incomplete capture of these variables among patients and an underestimation of these diagnoses; primary care data could not be collected by administrative databases used in the present analysis. Moreover, the data on the therapies indicated for IBD were taken without discriminating the role of conventional regimens and biological drugs on disease course. Lastly, in the cost and healthcare utilization analysis, a limitation could be that we cannot account for visits and expenses outside the national healthcare system, and we did not account for indirect costs.

Conclusion

The present real-world data analysis has shown for the first time that iron therapy supplementation in anemic IBD patients could be associated with beneficial effects in terms of IBD disease progression and healthcare resource utilization restraining in the Italian clinical setting. In addition, the high quote of undertreated anemic patients found among the real clinical practice, suggests that there is still an unmet need among IBD patients for treatment optimization.

Acknowledgements

None.

Conflicts of interest

CSL Vifor funded the study report, the contents of which are given in this manuscript, produced by Vifor Pharma Group and CliCon S.r.l. Benefit Society. The opinions expressed here are those of the authors and not necessarily those of the funders. The agreement signed by CliCon S.r.l. and CSL Vifor does not create any entity, joint venture or any other similar relationship between the parties. CliCon S.r.l. is an independent company. Neither CliCon S.r.l. nor any of its representatives are employees of CSL Vifor in any capacity. Antonio Ramirez de Arellano Serna and Neige Teldja Morin are employees of CSL Vifor. Ioannis Koutroubakis has served as advisory board member for Abbvie, Genesis, Janssen, MSD, Pfizer, Shire, Takeda and Vifor; Speaker for AbbVie, Genesis, Janssen, MSD, Vianex and Takeda; research support Abbvie, Vianex and Ferring. For the remaining authors, there are no conflicts of interest.