https://orcid.org/0000-0003-2634-9668
Abstract
The prevalence of inflammatory bowel disease (IBD) has been increasing worldwide, causing high impact on the quality of life of patients and an increasing burden for health care systems. In this systematic review, we reviewed the literature concerning the direct costs of Crohn’s disease (CD) for health care systems from different perspectives: regional, economic, and temporal.
We searched for original real-world studies examining direct medical health care costs in Crohn’s disease. The primary outcome measure was the mean value per patient per year (PPY) of total direct health care costs for CD. Secondary outcomes comprised hospitalization, surgery, CD-related medication (including biologics), and biologics mean costs PPY.
A total of 19 articles were selected for inclusion in the systematic review. The studies enrolled 179 056 CD patients in the period between 1997 and 2016. The pooled mean total cost PPY was €6295.28 (95% CI, €4660.55-€8503.41). The pooled mean hospitalization cost PPY for CD patients was €2004.83 (95% CI, €1351.68-€2973.59). The major contributors for the total health expenditure were biologics (€5554.58) and medications (€3096.53), followed by hospitalization (€2004.83) and surgery (€1883.67). No differences were found between regional or economic perspectives, as confidence intervals overlapped. However, total costs were significantly higher after 2010.
Our review highlighted the burden of CD for health care systems from different perspectives (regional, economic, and temporal) and analyzed the impact of the change of IBD treatment paradigm on total costs. Reducing the overall burden can depend on the increase of remission rates to further decrease hospitalizations and surgeries.
Introduction
Inflammatory bowel disease (IBD), including Crohn’s disease (CD) and ulcerative colitis (UC), affects patients of all ethnicities and geographic regions.1 Its prevalence has been growing worldwide, with 1 million patients in the United States and 2.5 million in Europe.2 In particular, CD is characterized by transmural inflammation of the intestine and can affect all the gastrointestinal tract, from the mouth to the perianal area. Its prevalence and incidence vary according to geographic region, ethnicity, and environmental factors.3
Both CD and UC are remitting and relapsing conditions that impair the quality of life of patients, with limitations in terms of career progression, relationships, and high socioeconomic consequences.4,5
The burden of CD is not restricted to patients. In fact, CD demands continuous medical care and monitoring provided by specialized care teams, including professionals who coordinate diagnosis and monitoring through tailored pharmacological interventions and other strategies like hospitalization and surgery, with a high impact on health care systems’ budget and organization of health care systems.6–9 Moreover, the introduction of biological agents and, more recently, the emergence of new small drug molecules and biosimilars revolutionized both CD and UC treatment,10–12 which shifted from symptom control to treat-to-target algorithms. However, patients can now see their conditions stabilized more rapidly, with increased odds of achieving remission, lower surgery rates, and higher survival rates.13
Considering all the characteristics of CD and the recent changes regarding its management and treatment, it is essential to evaluate the costs of CD for health care systems to optimize the available options and assess their impact on the overall impact burden of the disease.
Several systematic reviews were published on the costs of CD and UC, mainly centered on indirect costs and pediatric populations.14–19
This study intends to fill the knowledge gap by focusing on the direct costs of CD through different perspectives (regional, economic, and temporal) while discussing the management of CD through pharmacological treatments, surgery, and hospitalization.
Materials and Methods
Search Strategy
This study was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).20 MEDLINE (via PubMed), Scopus, and ISI Web of Science databases were searched from inception to July 14, 2020, with the following keywords or medical subject heading (MeSH) terms: Inflammatory Bowel Diseases(Mesh) or Crohn Disease (Mesh) or Colitis, Ulcerative (Mesh) or Crohn or crohn disease or colitis or ulcerative colitis and Costs and Cost Analysis (Mesh) or Health Care Costs (Mesh) or burden or cost or econom∗ or charge. No publication date or language restrictions were applied at this stage. To ensure that all relevant articles were identified, we manually searched reference lists.
Eligibility Criteria
Using the PICO model, we searched for original real-world studies examining direct medical health care costs in IBD as primary or secondary outcomes (through mean costs per patient per year), regardless of patient’s age. We did not consider any limit on intervention or comparison in this stage.
The exclusion criteria were (1) randomized controlled trials (RCTs), systematic reviews, review articles, animal and in vitro studies, guidelines, and editorials; (2) studies focusing on indirect costs, patient-reported costs, comparative effectiveness, cost-effectiveness, cost-utility, and other economic models or Markov analyses; (3) studies addressing the economic burden of IBD without providing the following precision metrics: mean, standard deviation, standard error, or confidence intervals; (4) studies not providing currency value of the costs; (5) studies with nonextractable data; and (6) studies not written in the English language.
Study Selection and Data Collection
Titles and abstracts were thoroughly analyzed, and studies that did not fulfill the eligibility criteria were excluded. Next, the full texts of the remaining studies were evaluated to determine their inclusion or exclusion but only regarding Crohn’s disease. The articles that, in the opinion of 2 authors (M.S. and C.A.), met the inclusion criteria were compared, and disagreements were solved by dialogue and consensus. From each selected study, the following information was collected: authors, publication year, cohort’s region/country of origin, publication type (full-text or abstract), study design, study period, number of patients enrolled, age category, mean age, gender, economic perspective, year of economic evaluation, currency, and costs and their category (total direct, hospitalization, surgery, medication, and biologics). These categories were attributed according to what was reported by the authors of the individual studies.
Data Assessment and Adjustment
Our primary outcome measure was the mean value per patient per year (PPY) of total direct health care costs for CD. Secondary outcomes comprised hospitalization, surgery, medication (all CD-related medication, including biologics, were assessed in this category), and biologics mean costs PPY. Outcomes were analyzed through 3 perspectives: regional (Europe, North America, Other), economic (national health service, third party payer), and temporal (before 2010, 2010 and afterward). Regarding the latter perspective, longitudinal studies reporting study periods embracing both categories were adjudicated to the one pertaining to most years disclosed.
Only studies presenting mean values PPY of direct costs, number of patients, and at least standard deviation, standard error, or confidence interval were eligible. Conversions between standard deviation, standard error, and confidence interval were performed following the Cochrane Handbook for Systematic Reviews of Interventions.21
If the study did not report global mean costs, we performed weighted averages for each analyzed group (eg, per age category, year, or disease severity).
In order to standardize different years and local currencies, reported costs were converted from the local currency in the reported year to euros (€), according to the Eurostat dataset22; converted costs were inflated into 2020 €.23
Quality Assessment
The methodological quality of the studies was based on the interpretation of the Larg and Moss checklist by Kinchin and colleagues.24 The authors assigned a global quality score to each study derived as a proportion of “yes” answers out of a total of 20 questions in the checklist.
Statistical Analysis
The main data analyzed in this study was the mean value PPY of direct health care costs and their categories (total, hospitalization, surgery, CD-related medication, and biologics).
For the cost analysis, we performed weighted pooled averages of the mean costs PPY reported in each study and their confidence intervals. Values were multiplied by each study’s sample size and divided by the sum of the sample sizes. Raw values of confidence intervals were converted into log-transformed values to prevent a negative lower limit.
Subgroup analyses were conducted to evaluate the impact of confounders of the variables: regional, economic, and temporal perspective. All analyses were executed using R software. Statistical significance was considered when confidence intervals did not overlap.25,26
Results
Literature Search and Study Selection
The electronic database search yielded 10 313 records (3269 records in MEDLINE, 2806 in Scopus, and 4238 in ISI Web of Science); the manual search did not identify additional studies. After the removal of duplicates (n = 4265), 6048 records remained, 5601 of which were excluded. The remaining 447 were assessed for eligibility. Of those, 323 addressed CD, although 304 were excluded. A total of 19 articles were selected for inclusion in the systematic review (Figure 1).
Flow diagram of the studies’ selection and data collection process.
Characteristics of Included Studies
The characteristics of the included studies are summarized in Supplementary Table 3. With the exception of one,27 all studies were full-text.28–46 Although all studies were published from 2000 onwards, the study period ranged from 1997 to 2016. Most pertained to Europe,27,31,32,34–36,38,40,41,44 7 to North America,28,29,33,37,39,42,43 and 3 to Africa,46 Asia,45 and Oceania.30 All studies were observational in nature, 14 of which were retrospective.
Overall, the studies enrolled 179 056 CD patients, varying from 4538 to 145 610 patients.37 The studies appear to have included adult patients; 5 presented data on all ages,30,33,35,37,43 and 6 were not clear when disclosing age categories.29,34,36,40,42,46 The majority reported the female gender as more prevalent.29–31,33,34,37–40,42–44
Regarding the economic perspective, 11 studies evaluated costs from the national health care system viewpoint,29–32,34,35,38,40,41,44,45 7 from a third-party payer,33,36,37,39,42,43,46 and 1 was unclear.27 Additionally, 9 studies reported data concerning the time period prior to 2010,27,29,31,32,34,39,40,42,43 and 10 from 2010 onwards.30,33,35–38,41,44–46
When assessing the outcomes, total direct health care costs were estimated in 17 studies,29–38,40–46 hospitalization costs in 13,29–32,34–36,38–40,42,43,45,46 surgery costs in 8,29–32,35,43,45,46 CD-related medication costs in 15,27,30,31,39–46 and biologics costs in 9.27,35,38,39,41–45
Cost Estimates
Total direct costs
The total direct mean costs PPY of CD was reported in 17 studies (Supplementary Table 3). The pooled mean cost PPY was €6295.28 (95% CI, €4660.55-€8503.41; Supplementary Table 1).
Regarding subgroup analysis, no effects were found for regional or economic perspectives (Supplementary Table 1).
Nonetheless, we observed that in regions apart from Europe and North America, total direct mean costs PPY were about 60% higher (other: €8539.29 95% CI, €7554.86-€9652.01]; Figure 2A), but without statistical significance, as confidence intervals overlap. From the perspective of the payer, even though total direct mean costs PPY of third-party payers were about 50% higher than those of national health services (€8014.71 95% CI, €5026.64-€12 779.03]; Figure 2B, Supplementary Table 1), the differences were not statistically significant between groups.
Total direct health care costs. A, Pooled mean costs per patient per year per region (n = 17). B, Pooled mean costs per patient per year per economic perspective (n = 17). C, Pooled mean costs per patient per year per temporal perspective (n = 17).
However, when comparing temporal perspectives, we found that studies performed after 2010 presented total direct costs significantly higher than before 2010 (€8611.56 95% CI, €6213.64-€11934.88] vs €4190.09 95% CI, €2975.94-€5899.59]; Figure 2C, Supplementary Table 1).
Hospitalization costs
The pooled mean hospitalization cost PPY for CD patients was €2004.83 (95% CI, €1351.68-€2973.59; Supplementary Table 1). This outcome was assessed in 13 studies and ranged from €908.80 (95% CI, €648.00-€1274.56) to €10 721.15 (95% CI, €9876.16-€11 638.43).
Because confidence intervals slightly overlap, statistically significant differences were not found in the analysis by regional, economic, or temporal perspectives (Figure 3). The pooled mean hospitalization costs PPY were lower in Europe (€1257.43, 95% CI, €960.69-€1645.82) compared with the rest of the world (North America, €3138.18, 95% CI, €1364.63-€7216.76; other, €2701.06, 95% CI, €1165.65-€6258.94; Figure 3A, Supplementary Table 1).
Hospitalization costs. A, Pooled mean costs per patient per year per region (n = 13). B, Pooled mean costs per patient per year per economic perspective (n = 13). C, Pooled mean costs per patient per year per temporal perspective (n = 13).
As for the economic perspective, pooled results from the perspective of national health systems (€2261.44, 95% CI, €1223.23-€4180.83) were higher than those from third-party payers’ viewpoint: €1648.17 (95% CI, €1226.64-€2214.55; Figure 3B, Supplementary Table 1).
Finally, pooled mean hospitalization costs PPY were similar before and after 2010 (before 2010, €1904.52, 95% CI, €1092.11-€3321.25; after 2010, €2172.21, 95% CI, €1224.24-€3854.23).
Surgery costs
From the included studies, 8 evaluated the mean costs PPY of surgery in CD. The pooled mean cost PPY was €1883.67 (95% CI, €761.39-€4660.14; Supplementary Table 1).
Surgery costs were also lower in Europe (€660.17, 95% CI, €467.68-€931.89) compared with other regions (North America, €3989.49 95% CI, €294.69-€54 009.04; other, €2981.41, 95% CI, €711.83-€12 487.27), but without statistical significance, as confidence intervals overlap (Figure 4A, Supplementary Table 1).
Surgery costs. A, Pooled mean costs per patient per year per region (n = 8). B, Pooled mean costs per patient per year per economic perspective (n = 8). C, Pooled mean costs per patient per year per temporal perspective (n = 8).
Regarding the economic perspective, pooled mean costs from the perspective of national health systems (€2123.29, 95% CI, €637.69-€7069.85) were higher in comparison with those of the third-party payer subgroup (€1068.33, 95% CI, €927.68-€1230.31) but without statistical significance (Figure 4B, Supplementary Table 1).
As for the temporal analysis, surgery pooled mean costs were slightly lower before 2010 compared with after 2010 (before 2010, €1679.25, 95% CI, €373.05-€7558.94; after 2010, €2103.01, 95% CI, €618.82-€7146.94), also without statistical significance (Figure 4C, Supplementary Table 1).
CD-related medication costs
This outcome was evaluated in 15 studies; all analyzed CD-related medication mean costs PPY in CD (including biologics). The pooled mean costs were €3096.53 (95% CI, €1736.91-€5520.44; Supplementary Table 1). Statistical significant differences were not found among subgroups, as confidence intervals overlapped.
Medication costs were about 3 times higher in North America (€9779.98, 95% CI, €3275.14-€29 204.26) than in the Europe subgroup (€3079.50, 95% CI, €1736.46-€5461.30) and about 10 times higher than in other subgroup (€984.57, 95% CI, €285.74-€3392.53; Figure 5A, Supplementary Table 1).
CD-related medication costs. A, Pooled mean costs per patient per year per region (n = 15). B, Pooled mean costs per patient per year per economic perspective. One study did not report its economic perspective (n = 14). C, Pooled mean costs per patient per year per temporal perspective (n = 15).
From the economic perspective, national health systems yielded lower pooled mean costs compared with third-party payers (€2267.26, 95% CI, €1375.55-€3737.01 vs €4032.49, 95% CI, €1007.76-€18 368.85; Figure 5B, Supplementary Table 1).
Additionally, pooled mean costs were approximately 69% higher before 2010 (€4097.74, 95% CI, €1568.61-€10704.69) than afterward (€2422.12, 95% CI, €1203.87-€4873.16; Figure 5C, Supplementary Table 1).
Biologics costs
The pooled mean costs PPY of biologics in CD were €5554.58 (95% CI, €2533.52-€12 178.09; Supplementary Table 1). This outcome was reported in 7 studies and varied from €901.88 (95% CI, €287.00-€2834.11) to €24 567.39 (95% CI, €23 460.20-€25 726.84). Because confidence intervals overlap, we found no statistically significant differences among perspectives.
Regarding regions, biologics costs were highest in North America (€9158.65, 95% CI, €1526.70-€54 942.63) and lowest in the other subgroup (€901.88, 95% CI, €287.00-€2834.11; Figure 6A, Supplementary Table 1).
Biologics costs. A, Pooled mean costs per patient per year per region (n = 9). B, Pooled mean costs per patient per year per economic perspective. One study did not report its economic perspective (n = 8). C, Pooled mean costs per patient per year per temporal perspective (n = 9).
In addition, national health systems presented about 2.5 times lower biologics costs than third-party payers (€3666.47, 95% CI, €1428.12-€9413.06 vs €9158.65, 95% CI, €1526.70-€54 942.63), but the difference was not statistically significant, as confidence intervals overlap (Figure 6B, Supplementary Table 1).
Lastly, biologics costs were about 2.5 times higher before 2010 (€9270.83, 95% CI, €2612.52-€32898.64) compared with after 2010 (€3666.47, 95% CI, €1428.12-€9413.06; Figure 6C, Supplementary Table 1).
Quality of Studies and Publication Bias
The scores obtained with the Larg and Moss checklist ranged from 30%27 to 100%,29,37,43 with a mean (± standard deviation) of 90.5% ± 15.4% (Supplementary Figure 1, Supplementary Table 2). In general, the highest scores were obtained for the questions, “What was the motivation and perspective of the study?” “Were all relevant, nontrivial cost components and their stakeholders identified?” “Were necessary timeframes specified?” “Did the analysis address the study question?” and “Was a range of estimates presented?” (Supplementary Table 2).
Discussion
In this systematic review, we reviewed studies presenting results for the direct health care costs of CD, including total direct costs and those related to medications (and biologics in particular), hospitalizations, and surgeries. Given the chronic nature of this condition and its increasing prevalence worldwide, the obtained results illustrate the high burden of IBD on health care systems, with a calculated mean total cost PPY of €6295.28, with statistically significant differences across time periods. On the other hand, no differences were found among regions or payers.
Considering the tendency for the reduction of mortality among IBD patients (as a result of aggressive medical therapy47,48) and the overall aging of the world population, health care systems should be prepared to provide continuous medical care to these patients for more extended periods.
These results are broadly in line with the evidence that has been reported for another chronic immune-mediated disorder, rheumatoid arthritis (RA), in a recent systematic review, as mean annual RA-related direct costs ranged from $3723 to $11 587 per patient.49 In contrast, compared with multiple sclerosis (MS) in the United States (US), the mean annual total health care cost per patient was reported to be $62 500 in 2015 for commercially insured MS patients.50
Our systematic review revealed that the major contributors for the total health expenditure in IBD were biologics (€5554.58) and CD-related medication (€3096.53), followed by hospitalization (€2004.83) and then surgery (€1883.67). This trend is in agreement with previous reports, which show the impact of the treatment paradigm change of IBD on the economic burden of the disease.48,51,52 Moreover, these results illustrate the current importance of biological therapy in the current treatment of IBD.
They also show that even though biologic costs are lower after 2010 (although without statistical significance), perhaps owing to the advent of the biosimilar drug, hospitalization and surgery costs do not seem to be declining, which align with earlier studies reporting little association between biologic use, hospitalization, or surgery rates at the population level.53,54
Nevertheless, we agree that the previously reported increase of the total burden of the disease48,51,52,55 may not only be due to the generalized use of biologics and biosimilars but also to an increase in emergency room and outpatient visits.
Additionally, despite not being the focus of this review, we must highlight the significant role of indirect costs on the overall cost burden of IBD, as was recently portrayed in a systematic review of 18 observational studies by Constantin and colleagues in which indirect costs accounted for 35% of total UC costs.16 An earlier study also reported that indirect costs accounted for approximately 33% of total UC costs in the United States and more than 50% in Europe.15
Overall, the costs associated with hospitalization, medications, and biologics were higher in North America than in Europe and other regions. This tendency may be related to the financing policies of the US health system, which rely on third-party payers and are in agreement with previous data in which costs of hospitalization, physicians, and clinics were referred as major contributors for the overall health expenditures in the United States.56 The drawbacks and limitations of these systems have been debated in the United States for decades, and several stakeholders have been demanding a restructuring that would decrease the overall health care costs.56 However in our study, the costs allocated to third-party payers were not statistically different from those attributed to national health care systems, indicating that there might be other sociocultural factors affecting the higher costs for hospitalization, medication, and biologics in North America. In this setting, we anticipate that diet, general health habits, and comorbidities might significantly contribute to this trend, as they can influence the patient journey, the therapeutic options, and the odds of achieving remission.56–58
This study presents some limitations. The representativity of our sample is low because with our inclusion criteria, from thousands of publications, we could include only 19 studies. However, systematizing the available data with accuracy demanded specific and rigid criteria based on mean costs PPY. With the selected search strategy, we can guarantee that the included studies provide data obtained under the same accounting system, increasing the robustness of the results. The study also presents the disadvantages of most systematic reviews because some studies might have been missed by the selected search expressions. Additionally, it does not portray the actual global economic burden of CD, narrowing the potential perspective of this analysis, as it is focused solely on the direct costs of the disease, which is in itself an issue of complex nature and prone to oversimplification; however, it is more extensively described in the literature and thus easier to synthesize.
In conclusion, our systematic review highlighted the burden of CD for health care systems worldwide from different perspectives (regional, economic, and temporal) and analyzed the impact of the change of IBD treatment paradigm on the costs per patient. In this context, the reduction of total costs may depend on higher disease remission rates to further decrease hospitalizations and surgeries. This process may be already on course due to the increasing use of biologics and may also benefit from a more generalized use of biosimilar drugs.
Patients and health care systems would also benefit from frequent re-evaluations of CD clinical management in terms of efficacy of all the therapeutic approaches. This would allow for the adjustment of treatments to achieve better results with lower costs, consequently reducing the economic burden of CD to health care systems and society in general.
Abbreviations
- CD
Crohn’s disease
- CI
confidence interval
- IBD
inflammatory bowel disease
- MEDLINE
Medical Literature Analysis and Retrieval System Online
- MS
multiple sclerosis
- PICO
population, intervention, comparison, outcome
- PPY
per patient per year
- PRISMA
Preferred Reporting Items for Systematic Reviews and Systematic review
- RA
rheumatoid arthritis
- UC
ulcerative colitis
Acknowledgments
The authors thank Paula Pinto, PharmD, PhD (PMA, Pharmaceutical Medicine Academy), for providing medical writing and editorial assistance. M.S. acknowledges “Fundação para a Ciência e Tecnologia (FCT),” Portugal under grant number PD/BD/142890/2018; PhD Program in Clinical and Health Services Research (PDICSS).
Author Contributions
M.S. was involved in data acquisition, analysis, interpretation, and manuscript drafting; C.C.D. was involved in data analysis, interpretation, and manuscript revision; C.A. was involved in data analysis, interpretation, and manuscript revision; F.M. coordinated the study’s conception and design and was involved in data interpretation and manuscript revision. P.M., R.G., D.C., F.P., L.C., and P. L. were involved in data interpretation and manuscript revision. All authors approved the final version of the article.
Funding
This work was supported by the Portuguese Study Group of Inflammatory Bowel Disease (GEDII).
Conflicts of Interest
F.M. served as a speaker and received honoraria from Merck Sharp & Dohme, Abbvie, Vifor, Falk, Laboratórios Vitória, Ferring, Hospira, and Biogen. The other authors have no conflict of interest to disclose.
Data Availability
The data underlying this article will be shared at reasonable request to the corresponding author.
References
