As shown in the PRISMA flow diagram (Fig. 1), 1055 records were identified through database searches. After duplicates removal and the established selection/inclusion criteria were applied, 45 publications (33 full-text articles and 12 abstracts) were included in the synthesis.

PRISMA flow diagram showing the selection process of the included publications

The characteristics of each publication are detailed in Table 1. Of note, the data described in the 45 publications are derived from 38 different studies. Out of 45 publications, 34 targeted patients with IBD, with different levels of granularity in reporting the data (IBD overall and/or separately for UC and CD), while five focused exclusively on CD and six on UC. A summary of the main characteristics of these studies is presented in Table 2.

Incidence Fourteen publications reported data on IBD incidence [overall IBD (n = 10), CD (n = 13) and UC (n = 14)], 12 of them in adults (including studies involving patients of all ages) and two in paediatrics. The incidence of IBD, CD and UC varied greatly by study, region and year, ranging from 9.6 to 44.3 per 100,000 inhabitants (Fig. 2) [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29]. The incidence ranges for CD and UC were 4.6 to 18.5 and 3.4 to 26.5 per 100,000 inhabitants, respectively.

Incidence of IBD according to year of analysis: overall (i), CD (ii) and UC (iii) according to year

The most recent data at the national level identified in this review estimated adult overall IBD incidence to be 16.2 per 100,000 inhabitants, 7.4 for CD and 8.1 for UC in 2017, with higher rates reported in Asturias and Navarra (Fig. 3).

Incidence of adult IBD at regional level in 2017

Two studies conducted at the regional level (in Castilla-La Mancha and Cataluña) described an increase in IBD incidence between 1.5- and twofold in the period from 2000 to 2016 [16, 25] (Fig. 2i). A similar trend was reported for CD and UC at the national and regional levels, except in Andalucía, where incidence remained more stable (Fig. 2ii, iii) [16, 18, 25].

Prevalence Four publications reported data on adult IBD prevalence [overall IBD (n = 3), CD (n = 2) and UC (n = 3)]. Paediatric prevalence data were not available. Adult prevalence ranged from 79.8 to 545.3 per 100,000 inhabitants for IBD, from 37.2 to 191.4 for CD and from 41.5 to 354.0 for UC (Fig. 4).

Longitudinal trend of IBD, CD and UC prevalence in two Spanish regions

The most recent prevalence estimate at the national level identified in the search was 88.7 per 100,000 inhabitants [26] in 2011. However, other studies with a regional scope reported a higher prevalence (> 200) in the same year [16, 25], and an increasing trend over time (Fig. 4).

According to sex, data from Castilla-La Mancha showed a higher prevalence of UC in men than in women (115.39 vs. 84.54; p = 0.015), with no significant differences in CD [25].

Mortality Two publications reported IBD-associated mortality data [overall IBD (n = 2), CD (n = 2) and UC (n = 2)] [16, 30].

Over the period 2006–2015, CD and UC mortality rates of up to 6.0% and 3.0%, respectively, were reported [30]. More recently, an increase in mortality rates in Spain (expressed per 1000 inhabitants) was observed from 2011 to 2016 for IBD (14.7–18.6; 27% increase), CD (12.5–17; 36% increase) and UC (15.7–19.4; 24% increase) [16]. Compared to patients without IBD, the age and sex-adjusted odds ratio for death due to IBD were significantly higher for patients with CD (RR 1.63; 95% CI 1.39–1.89) and UC (RR 1.22; 95% CI 1.11–1.36) [16].

Seventeen publications reported data on the age of IBD onset [overall IBD (n = 11), CD (n = 14) and UC (n = 12)], 15 of them in adults and 2 in the paediatric population.

Disease onset predominantly occurs between 30 and 40 years of age in adult patients and seems to be more delayed for UC than CD (Table 3). Interestingly, the opposite trend was described in the two publications focused on the paediatric population, with an earlier diagnosis for patients with UC than for patients with CD (11.5 vs. 12.7 years; p < 0.001) [28].

The clinical data hereby presented has been recorded at different times: 9 studies collected these data at diagnosis, 5 at study entry and 15 at any time in the course of the disease/study (Table 4).

Gastrointestinal manifestations are present in 94.0% of patients with CD and 89.0% of patients with UC [19]. Despite this, they have only been described in one publication for UC, with rectal bleeding (88.8%), diarrhoea (80.0%), pain (69.1%) and rectal urgency (59.3%) being the most common signs/symptoms [31]. Weight loss in 34% of adults and 78% of children and vomiting in 3.8% of adults and 32% of children have been reported in CD [32].

On the other hand, EIMs were described in 16 publications, with their presence ranging from 7.0% to 28.7% for overall IBD (Table 4). A higher prevalence of EIM in patients with CD compared to patients with UC was observed in five out of seven publications with available data [19, 30, 33,34,35]. The most common EIM was osteoarticular manifestations, reported in over 10% of patients for both CD and UC.

Between 13.5% [35] and 26.6% [36] of adult patients with IBD in the studies had at least one other immune-mediated inflammatory disease (IMID)

Psychiatric comorbidities were frequently reported, with depression described in up to 20% and anxiety in up to 11% of patients with CD and UC [19, 30, 33,34,35].

The most frequently diagnosed IMID and non-IMID comorbidities are described in Table 5.

Although the overall risk of cancer did not significantly increase in patients with IBD [15, 37], the relative risk of developing some types of cancer has been reported: urothelial carcinoma (RR 5.23, 95% CI 1.95–13.87), appendiceal mucinous cystadenoma (RR 36.6, 95% CI 7.92–138.4), small intestine carcinoma (RR 13.1, 95% CI 1.82–29.7) and rectal carcinoid (RR 8.94, 95% CI 1.18–59.7) [15]. Two studies also evaluated the possible effect of thiopurines on the risk of extracolonic cancer [15] and overall neoplasm [37] but a clear association between variables was not found. Thus, it could not be concluded whether the risk of malignancy was attributable to a given drug, its combination with another, the time under treatment, doses or the disease itself. No CD- and UC-specific data were available.

Regarding the location of the CD lesion, heterogeneous data were observed according to the Montreal classification, which was adopted by all the publications (see description of mentioned disease classifications in the footnote of Table 4). However, L1 occurs more frequently in the adult and elderly population than in the paediatric population, whereas L3 shows the opposite trend (Table 4). Moreover, the paediatric population has been found to have the highest frequency of ileal involvement (L1) in 13–17-year-olds (30.3%) as compared to under 5 years (25%) and 6–12 years (22.6%) (p < 0.001) [27]. Disease behaviour in CD was described in 19 publications, with inflammatory type (B1) being the most frequently reported (14 out of 19 publications). Interestingly, the frequency of perianal disease was higher in the paediatric than in the adult population (Table 4).

Heterogeneous data were also observed in UC extent among the studies included in our review. Nonetheless, in the adult population E2 was the most frequent localization compared to E1 and E3. On the other hand, most studies in the paediatric population showed a higher prevalence of E3 than those carried out in the adult population (Table 4). In addition, E3 was significantly predominant in children under 5 years compared to 6–12-year-olds and 13–17-year-olds [27]. Disease severity in UC was only described in one publication [37], with most patients having S1 (51%) or S2 (41%).

Nine of the publications reviewed assessed HRQoL in patients with IBD [overall IBD (n = 2), CD (n = 5) and UC (n = 6)] [33, 38,39,40,41,42,43,44,45]. Data for the paediatric population were not available.

Patients with IBD had lower HRQoL (as measured with the SF-36) than the reference values of the general population [40]. In addition, two studies in patients with IBD reported an association between high levels of anxiety, depression and stress (measured with the Hamilton Depression Rating Scale (HAM-D), Hospital Anxiety and Depression Scale (HADS), and Perceived Stress Scale (PSS) questionnaires) and low levels of HRQoL and more significant symptomatology, with no differences between CD and UC [40, 42]. Another study showed that patients with CD reported higher levels of sexual dysfunction than healthy controls (35% vs. 12%, p < 0.08) [45], with significant differences in erectile function, orgasm, sexual desire and global satisfaction (p < 0.05) [39, 45]. Among those patients who felt that intimacy had worsened because of IBD, fatigue was the leading complaint in men and women [39].

In general, no significant differences in HRQoL between patients with CD and UC have been described [38, 40, 42]. However, two studies reported a significantly poorer HRQoL in patients with CD compared to patients with UC, measured with the Inflammatory Bowel Disease Questionnaire (IBDQ-32): IBDQ-32 mean score 155.4 (SD 42.7) vs. 180.3 (SD 32.4), p = 0.005 [41]; IBDQ-36 functional domain 44.8 (39.9–47.6) vs. 46.9 (45.5–49.0), p = 0.02 and social affectations 39.6 (36.0–40.2) vs. 39.6 (39.6–40.8), p = 0.04 [38].

Women with IBD had poorer HRQoL than men [IBDQ-32 score 152.5 (SD 43.3) vs. 180.6 (SD 31.4), p = 0.001] [41], and a higher impact of UC on sleep quality and higher levels of anxiety and depression were reported in women than in men (p < 0.01) [43]. Data on the impact of CD on HRQoL in relation to sex were not available.

Other publications pointed out the role of age, EIMs, the presence of an exacerbation, the number of previous recurrences, disease activity and disease duration in HRQoL, suggesting a worse quality of life the more severe the disease is [38, 40, 43, 44].

The HRQoL among IBD treated patients was significantly improved in those who achieved remission after 1 year of biologic treatment while clinical activity decreased and normalization of HRQoL (IBDQ score > 209) was achieved in 74% of patients (67.4% CD vs. 100.0% UC, p < 0.05) [38].

Twenty-three publications reported qualitative and quantitative data on drug treatment patterns [overall IBD (n = 8), CD (n = 11) and UC (n = 9)]. Twenty-two targeted the general or adult population and one the paediatric population. However, specific percentages of use (quantitative data) are reported in only 17 of these publications (Table 6).

Considerable variability in treatment patterns was observed across studies. In general, from 1996 to 2018 aminosalicylates were the most frequently used treatments in Spain followed by immunomodulators and corticosteroids, while biological drugs were the least prescribed (Table 6). However, a significant increase in topical salicylates, systemic steroids, immunosuppressive drugs, and biologics and a reduction in topical steroids and oral aminosalicylates were reported between 1991 and 2011 [46]. The most recent data on the use of biologics at the national level derives from the ENEIDA registry, estimating that 25% of patients with CD were treated with biologics in 2016 [47], with no reported data for UC.

According to the IBD type, the use of biological treatment observed in the studies was more frequent in patients with CD (15.0–60.0%) than in patients with UC (6.9–36.0%) (Table 6). Moreover, data from the ENEIDA registry show that patients with CD had a higher risk of using immunosuppressants (HR 3.2 [95% CI 3.1–3.4]) and biological agents (HR 2.5 [95% CI 2.3–2.7]) than patients with UC [48].

The ENEIDA registry also reported that the use of immunomodulators and biologics was significantly higher in patients with childhood‐onset IBD than in those with adult-onset, [CD (85% vs. 66.2%, p < 0.001) and UC (56.1% vs. 28.3%, p < 0.001) and [CD (65% vs. 41.5%, p < 0.001) and UC (33% vs. 17.4%, p < 0.001), respectively] [48]. However, the median time from IBD diagnosis to the first biologic agent was similar in paediatric and adult‐onset patients (13 vs. 12 months, p > 0.05) [48].

Differences in treatment patterns between patients with familial and sporadic CD were also observed, showing a higher use of immunomodulators (79.9% vs. 63.1%) and biological therapy (54.4% vs. 38.6%) in the familial group [30]. Furthermore, in patients with IBD the presence of EIM was associated with a higher risk of using immunosuppressants or biological agents (1.2 [95% CI 1.1–1.3] and 1.7 [95% CI 1.6–1.7], respectively) [48].

Resource use Twelve publications reported data on resource use [overall IBD (n = 4), CD (n = 5) and UC (n = 6)] [16, 19, 26, 31,32,