The disease burden of human cystic echinococcosis based on HDRs from 2001 to 2014 in Italy

Background Cystic echinococcosis (CE) is an important neglected zoonotic parasitic infection belonging to the subgroup of seven Neglected Zoonotic Disease (NZDs) included in the World Health Organization’s official list of 18 Neglected Tropical Diseases (NTDs). CE causes serious global human health concerns and leads to significant economic losses arising from the costs of medical treatment, morbidity, life impairments and fatality rates in human cases. Moreover, CE is endemic in several Italian Regions. The aim of this study is to perform a detailed analysis of the economic burden of hospitalization and treatment costs and to estimate the Disability Adjusted Life Years (DALYs) of CE in Italy. Methods and findings In the period from 2001 to 2014, the direct costs of 21,050 Hospital Discharge Records (HDRs) belonging to 12,619 patients with at least one CE-related diagnosis codes were analyzed in order to quantify the economic burden of CE. CE cases average per annum are 901 (min—max = 480–1,583). Direct costs include expenses for hospitalizations, medical and surgical treatment incurred by public and private hospitals and were computed on an individual basis according to Italian Health Ministry legislation. Moreover, we estimated the DALYs for each patient. The Italian financial burden of CE is around € 53 million; the national average economic burden per annum is around € 4 million; the DALYs of the population from 2001 to 2014 are 223.35 annually and 5.26 DALYs per 105 inhabitants. Conclusion In Italy, human CE is responsible for significant economic losses in the public health sector. In humans, costs associated with CE have been shown to have a great impact on affected individuals, their families and the community as a whole. This study could be used as a tool to prioritize and make decisions with regard to a surveillance system for this largely preventable yet neglected disease. It demonstrates the need of implementing a CE control program aimed at preventing the considerable economic and social losses it causes in high incidence areas.


Methods and findings
In the period from 2001 to 2014, the direct costs of 21,050 Hospital Discharge Records (HDRs) belonging to 12,619 patients with at least one CE-related diagnosis codes were analyzed in order to quantify the economic burden of CE. CE cases average per annum are 901 (min-max = 480-1,583). Direct costs include expenses for hospitalizations, medical and surgical treatment incurred by public and private hospitals and were computed on an individual basis according to Italian Health Ministry legislation. Moreover, we estimated the DALYs for each patient. The Italian financial burden of CE is around € 53 million; the national average economic burden per annum is around € 4 million; the DALYs of the population from 2001 to 2014 are 223.35 annually and 5.26 DALYs per 10 5 inhabitants.

Conclusion
In Italy, human CE is responsible for significant economic losses in the public health sector. In humans, costs associated with CE have been shown to have a great impact on affected individuals, their families and the community as a whole. This study could be used as a tool to prioritize and make decisions with regard to a surveillance system for this largely PLOS

Introduction
Cystic echinococcosis (CE), caused by the larval stage of the taeniid tapeworm Echinococcus granulosus sensu lato, is an important zoonotic disease with a worldwide distribution. The World Health Organisation (WHO) included CE on the official list of the 18 Negleted Tropical Diseases. In addition to this group, the WHO has identified CE as belonging to a subgroup of seven endemic or ''neglected zoonotic diseases" (NZDs); these diseases are common where poverty, reliance on livestock or wildlife for social and financial capital, poor resilience, and the close proximity of people and their animals favor transmission [1] [2] [3]. Human CE remains highly endemic in pastoral communities, particularly in regions of South America, the Mediterranean littoral, Eastern Europe, the Near and Middle East, East Africa, Central Asia, China and Russia [4] [5] [6]. Domestic transmission of the infection relies on dogs as definitive hosts and a range of livestock ungulate intermediate hosts, mainly sheep and cattle [7] [8]. In the human host, the accidental oral ingestion of E. granulosus eggs may cause the development of cysts in many anatomic sites [9] [10].The initial phase of the primary infection is always asymptomatic. Small, well encapsulated, non-progressive or calcified cysts typically do not induce major pathology, and patients may remain asymptomatic for years or permanently [11]. The induction of morbidity depends on the number, size, and developmental status of the cysts (active or inactive), the involved organ, the localization of the cysts within the organ, the pressure of the cysts on the surrounding tissues and structures, and the defense mechanisms of the infected individual. Clinical signs may occur after a highly variable incubation period of several months or years [7] [12]. The disease represents a serious human and animal health concern that leads to significant economic losses derived from the costs of medical treatment, morbidity, life impairment and fatalities in human cases [13] [14]. CE is widespread in many regions of the world, affecting over 1 million people worldwide [15] [16]. Present estimates indicate that its global burden is about 184,000 DALYs (0.98 DALYs per case) [17]. The global financial burden of this disease in estimates of purchasing power parity is 4.1 billion international dollars annually, of which 46% is due to human treatment and morbidity and 54% is associated with animal-health costs [18] [19].
The reports from the EFSA and the ECDC highlight that a surveillance system on human CE is absent in Italy. As a result, no official data are transmitted to European authorities [20]. Due to the lack of official data, the Italian National Reference Centre for Echinococcosis (CeNRE) analyzed Hospital Discharge Records (HDRs) drawn from the National Ministry of Health. A retrospective epidemiological study of HDRs for CE cases carried out between 2001 and 2012 in Italy indicated that in Italy human CE continues to be a significant public health problem, particularly in the Islands and in the South. The study revealed an increased risk of human CE in correlation with "rural areas with comprehensive development problems" and within areas where sheep breeding is widely practiced [21]. The aim of our study is to perform an analysis of human echinococcosis disease costs in Italy; in particular, to calculate the economic burden of hospitalization costs and treatment and to estimate the Disability Adjusted Life Years on cases from 2001 to 2014. Objective estimates of the burden of CE are important because its underestimation downgrades its relevance to policy-makers and funding agencies [22] [23].To our knowledge, this study was the first of its kind to be performed in Italy.

Ethics statement
In compliance with the current Italian legislation on privacy, it should be noted that the publication and or dissemination of HDR's data and their processing must take place only in aggregate form.

Study area
Italy is a country in South-Central Europe, occupying a peninsula and two big islands (Sicily and Sardinia) in the center of the Mediterranean Sea, with around 60 million inhabitants. Moreover, it is the 5 th country by population on the EU list (comprised of 28 countries) and 8 th on the 2015 World Bank ranking list of countries by their gross domestic product (GDP) [24]. In order to subdivide its territory, we adopted the Nomenclature of Territorial Units for Statistics (NUTS) 2010 classification (Eurostat, Regulation EC 1059/2003; European Parliament and of the Council, 2003b): a geocode standard for referencing the administrative divisions of countries for statistical purposes [25]. Italy is divided into 20 administrative units: the regions (NUTS 2). Regions are further grouped into five "macro regions" (NUTS 1). The Italian NUTS 1 are the North West (Valle d'Aosta, Liguria, Lombardia, Piemonte) with 16,131,000 inhabitants, the North East (Emilia-Romagna, Friuli-Venezia Giulia, Trentino-Alto Adige/Südtirol, Veneto) with 11,654,000 inhabitants, the Center (Lazio, Marche, Toscana, Umbria) with 12,071,000 inhabitants, the South (Abruzzo, Puglia, Basilicata, Calabria, Campania, Molise) with 14,168,000 and the Islands with 6,759,000 inhabitant (Sardegna and Sicilia) [26].

Hospital discharge records
The data used to perform the study on the economic and health burden of CE were obtained from HDRs. The HDRs data regarding the period from 2001 to 2014, in anonymous form and free of personal information, were provided to CeNRE by the Italian Department of Health. In Italy, Since January 1 st , 2009, the 2007 version of the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) for encoding diagnosis and procedures contained in the HDRs has been adopted as the standard.
The HDRs summarize the medical records and classify diseases according to the ICD-9-CM and are divided by Disease Related Group (DRG), Major Disease Classification (MDC), type (medical or surgical) and description of DRG [27]. The HDRs contain an anonymous individual code for tracking patient's hospital admissions, discharges and readmissions. Each anonymous individual code identified one patient. Specifically, our database include: patient's admission and discharge dates, gender, age, domicile code, primary and secondary diagnoses codes, type of treatment (surgical or medical), length of stay in the hospital and mortality data. Data about hospitals include: regional and province code, department code, ward, type of hospitalization including ordinary hospitalization (OH) or day hospitalization (DH). OH refers to procedures that require an overnight hospital stay, while DH refers to procedures that require in-hospital treatment without an overnight stay. S1 Table describes all fields included in HDR schema.
The HDRs were converted to MySQL relational database management system (RDBMS) tables; calculations of cost elements were been performed as queries. The data, received as text file, were recorded in Microsoft Office Excel spreadsheets for better readability and in order to provide a graphic display of data and of grand total calculations. After integration of all information into the database, an extensive data checking and verification of completeness and compliance was evaluated in order to avoid the possibility of error and of reporting anomalies during calculation of direct costs. Baseline distribution of all variables included were evaluated in order to describe epidemiological situation during study period. All epidemiological analyses were performed using STATA software v.13; all test were two-side and p-value of 0.05 is considered statistically significant. We calculated the total number of records with echinococcosis related DRG code, subdivided by citizenship code. The CE patient was characterized using the individual anonymous code together with ICD-9 CE code at least once in the period of the study. For epidemiological purpose we defined as CE case the hospitalization (with CE in primary or secondary diagnosis) of a patient. To compute the Ministry of Health overall remuneration for CE, we identified CE hospitalizations as all records with CE in primary diagnosis. Furthermore the case fatality rate was calculated, using the discharge to death code of patients combined with CE primary diagnosis code.

Direct cost calculation
The HDR is an economic tool used to quantify and justify the costs incurred by public and private hospitals for hospitalizations and medical and surgical treatments.
Discharge diagnosis for Echinococcosis is identified according to Major Disease Classification (MDC) as MDC 18: Parasitic and systemic infectious diseases, the diagnosis code 122.0-122.9 Echinococcosis. In particular: 122.0 "Echinococcus granulosus infection of liver", 122.1 "Echinococcus granulosus infection of lung", 122.2 "Echinococcus granulosus infection of thyroid", 122.3 "Echinococcus granulosus infection, other", 122.4 "Echinococcus granulosus infection, unspecified", 122.5 "Echinococcus multilocularis infection of liver", 122.6 "Echinococcus multilocularis infection, other", 122.7 "Echinococcus multilocularis infection, unspecified", 122.8 "Echinococcosis, unspecified, of liver", 122.9 "Echinococcosis, other and unspecified". In Italy, reimbursements are detailed in Department of Health legislation concerning the remunerations paid out to hospital care services for acute cases as well as for rehabilitation and long-term care for post-acute and specialist outpatient care. From January 1997 to December 2012 it was regulated by the rate table of the Ministry decree of June 30 1997 on annex 1, while since January 2013 it has been updated with the annex of the decree of December 18 2008 amended on October 18 2012. The codes for reimbursement from 1997 to 2013 were: MDC 423 DRG 18 Type M (medical) "Diagnosis related to infectious and parasitological diseases", MDC 415 DRG 18 Type C (surgical) "Surgery for infectious and parasitological diseases"; since 2013 MDC 423 DRG 18 Type M (medical) "Diagnosis related to infectious and parasitological diseases" and MDC 578 DRG 18 Type C (surgical) "Surgery for infectious and parasitological diseases" [27].
The amount of government remuneration at the time of discharge is linked with primary diagnosis; therefore, only patients with primary diagnosis codes ICD9-CM 122.0-122.9 were considered for this computation. Payment distribution is further divided into: treatment type (surgical intervention or medical treatment), ordinary hospitalization (OH) (over 1 day and up to 80 days), ordinary hospitalization lasting one day or less due to the patient being transferred or deceased, and day hospital (DH). [28]. Since individual HDRs were available, to compute direct costs there was no need to group patients by age or by year of hospitalization. The total costs was subdivided by NUTS2 and by citizenship. Direct costs and number of hospitalizations over the 2001-2014 period were graphically represented, the Average Annual Cost (AAc) per 10 5 inhabitant, based on mean population in the study period was calculated [29].

Disability-Adjusted Life Years (DALYs) calculation
All Italian patients with CE diagnosis code were considered for DALYs calculation.
Using HDRs individual code we were able to extract each single patient's data through case history and to calculate the duration of the disease for each patient. The definition of Disability-Adjusted Life Years is: In this study, as in recent GBD studies the age weighting and discount were not used, in this way:

Results
This study delineates the economic picture of the costs of public health and the burden of this disease, calculated on the basis of HDRs, for the Human Cystic Echinococcosis in Italy from 2001 to 2014. First we analyzed the epidemiological situation, then we proceeded to calculate direct cost of hospitalizations and to the DALYs calculation. An analysis of 21,050 HDRs evidenced the presence of 12,619 patient (10,901 Italian patients and 1,718 foreign patients).The surgical patients amount to 3,634 and the medical patients were 8,967. The case fatality rate was 0,36% (SE 0.05%). The HDRs of patients with Italian citizenship and with discharge diagnosis code referred to Echinococcus multilocularis (ICD-9 code 122.5-122.7) are 903 related to 784 patients (Fig 1). These records can be recognized as misclassifications, with a diffuse distribution throughout the Italian territory and with the average regional frequency of 9.96% (±2.92 C.I. 95%) (Fig 2). These diagnoses can be defined errors because both official data and recent and past literature fail to demonstrate the presence of human AE in Italy [4]. Tables 1 and 2

Direct cost of hospitalization
The direct CE costs correspond to the overall remuneration provided by the national health services to hospitals. Human CE-associated economic losses that arise from diagnostic procedures, surgery and drug treatment, hospitalizations, periods of recovery, and fatalities are relatively easy to extrapolate from HDRs and enable the calculation of the direct costs of patients suffering from clinical echinococcosis. We considered 13,963 HDRs with primary CE diagnosis code of which 11486 HDRs related to Italian patients and 2477 to foreign patients. From

Discussion
Nowadays in Italy no CE surveillance system is in place. The Italian registry of CE (RIEC) was launched only in 2012. RIEC is a prospective multicentre registry of CE patients that have been visited since January 2012 in Italian health centers; data are voluntarily submitted to the registry [31]. RIEC was integrated in the European Register of Cystic Echinococcosis (ERCE), launched in October 2014 in the context of HERACLES project of the European Union Seventh Framework Programme (FP7/2007-2013) [32]. ERCE is a European prospective, observational, The burden of human cystic echinococcosis in Italy multicentre registry of patients with suspected or confirmed CE diagnosis. In Italy no CE screening studies were performed in the last years, either in adults or in schoolchildren [33]. Further national studies are needed in order to evaluate the prevalence of undiagnosed CE. Two out of five WHO collaboration centers for echinococcosis are in Italy, one at the University of Pavia, for Clinical Management of Cystic Echinococcosis, the newest one at Istituto Superiore di Sanità in Roma, finalized to the study of epidemiology, detection and control of cystic and alveolar echinococcosis (in humans and animals) [19]. Unfortunately, in the Italian context, the use of the adjective "neglected disease" for CE is particularly fitting for the current unavailability of extracting information from official database [20]. In the previous study conducted by National Reference Laboratory for CE (CeNRE) it was clear that HDR-based studies are a useful tool to evaluate the disease epidemiology, nevertheless it cannot replace a surveillance system implemented by official notifications [21]. Previous studies assessed the accuracy and the HDR's contribution in estimating prevalence of different diseases, comparing this data source with others (i.e. drug prescriptions, health-tax exemption, causes of death, cancer registries); depending on the disease, those studies demonstrated that the absolute contribution given by HDRs is between 8-72% of all disease cases [34] [35] [36]. In their paper, Rosso and Zanetti [37] underline the great diffusion of HDRs studies and the importance of HDRs information by pathology to use it as database; they also consider the HDRs as an appropriate tool for economical and programming studies. Since cooperation for the implementation of RIEC in 2012, the CeNRE intention has always been to support the effort of the two WHO collaborating centers in Italy in order to have an Italian epidemiological picture as accurate as possible of spread and disease distribution in animals and humans. However, in the use of HDRs there are some limiting factors, such as the ability of the physician to provide a correct diagnosis, and the unknown proportion of cases managed as outpatients by "watch and wait" approach according to the "Expert Consensus for the diagnosis and treatment of cystic and alveolar echinococcosis in humans" and hence unreported. Indeed, our study aims to evaluate the economic impact on the finances of Italian Public Health Service through the calculation of the total remuneration for The burden of human cystic echinococcosis in Italy the ICD-9 codes related to echinococcosis in the period 2001-2014 and providing first set of processed data, useful as starting point for comparison with data acquired in the future [38] [39] [21]. Finally, remarking again the underreporting and the high number of mistakes in the management of records, we aim to suggest, for the first time in Italy with the support of national economical data, the need for an improved ad hoc information flow for this neglected disease.

The data
The HDRs database is the only official database that covers all fourteen years, from 2001 to 2014, making it suitable for long-term studies and maximizing the chance of inclusion of undiagnosed or asymptomatic cases [34]. Data were extracted from all HDRs with at least one Cystic Echinococcosis diagnosis, either in the primary diagnosis or in one of the five possible secondary diagnoses. This allow for including many cases diagnosed by chance and many patient with developed disease prior to the study period. We did not consider asymptomatic patients since it is not possible assess their reimbursement costs, also they do not fit in to computation of DALYs because they do not qualify as "people living with disability". The estimation of HDRs underreporting in Italy will be better achieved after full development of National and European Registry of Echinococcosis that could allow for linkage studies aimed to quantifying accuracy of databases [40]. The burden of human cystic echinococcosis in Italy

Direct costs calculation
The considerations about burden of CE must not exclude an assessment of the economic impact of this disease upon the National Health System. The economic evaluation of the costs incurred in Italy for this disease was based on institutional economic repayment of the National Public Health System; the HDRs were provided by the Italian Ministry of Health. The total costs in the 14 years of the study amounted to around 53 million euro. The trend in costs was influenced by both the number of hospitalizations with CE primary diagnosis and the remuneration value fixed by the Health ministry. From January 1997 to December 2012 the reimbursements were regulated by the rate table ex DM 30/06/1997 annex 1, since January 2013 it has been updated by the annex of the law of December 18, 2008 amended on October 18, 2012. Updates of prices table increased remunerations up to 162% (e.g. surgical intervention from € 6,972.68 to € 18,314.00). In fact, as the graph suggests, costs followed a negative trend from 2001 to 2012 and a peak in 2013 (Fig 6). In their paper, Narra et al. found out that for 15 surgical patients treated in single center from 2008 to 2014, the surgical intervention related costs ranged from 5,874 to 23,077 per patient, the present study analysed 4239 surgical hospitalization of national database for 14 years, giving for a CE related surgical procedure an average cost of € 8,152; considering the update of remuneration prices across the study period, The burden of human cystic echinococcosis in Italy these findings are not in contradiction [41].This study show that the geographical breakdown of costs were strictly related to geographical distribution of disease conducted in a previous study with peaks in NUTS 1 Islands and Southern Italy and less financial commitment in the North [21]. There was an even more marked difference in CE costs related to the population in the various Italian regions (Fig 3, Fig 4). In NUTS 1 the ratio of Italian and foreign patients is different, with a substantial balance of hospitalizations among Italian and foreign patients in northern Italy and on the other hand with a high prevalence of hospitalizations of Italian patients in the South and in the Islands. This reflects the distribution of foreigners in the Italian

DALYs
The application of the DALYs formula (DALYs = YLL+YLD) revealed how in Italy this disease is in proportion rarely mortal (32% YLL), in fact this study showed a low case fatality rate (0.36%) and a high average age of fatal cases (63.61 years). The burden is mainly influenced by morbidity and duration (68% YLD). European DALYs median rate (95% uncertainty interval) per 100,000 inhabitants in 2010 reported by Torgerson et al. [44] was 0.8 (0.3-2); the Italian average annual DALYs in the study period per 100,000 inhabitants was 0.376 (SEM 0.035). Fig  7 represents geographical distribution by NUTS 2 of DALYs per 100,000 inhabitants in the study period. This confirm that Italy is part of the European CE endemic countries. Conducting an economic evaluation and a burden of disease study before the implementation of CE surveillance programs, should be the point of departure for a cost benefit study, that could stress the need of surveillance measures in high incidence areas [45].