Introduction

Acute and chronic hepatitis can be brought on by the hepatitis C virus (HCV), which increases the risk of cirrhosis, liver cancer, and death [1–2]. An estimated 3.9 million people in the European Union/Economic European Area (EU/EEA) are among the 57.8 million people with HCV [3] who are chronically infected worldwide. Due to the fact that chronic HCV infection is frequently asymptomatic, a significant number of infected individuals go undiagnosed [4]. The prevalence of HCV is unequally distributed throughout the population, with people living in prison (PLP), people who use drugs (PWUD), people living withHIV-positive (PLHIV), or people with certain chronic illnesses (e.g., patients undergoing hemodialysis, diabetics, or those who received blood or other products of human origin) being disproportionately affected [5]. At present, no studies have evaluated the effectiveness of population-level health policies implemented to increment HCV test and treatment coverages in Italy.

According to the available evidence [6–11], an estimated 398,610 people in Italy (1.7% of the population) have active HCV infection, with central regions having the highest prevalence (0.88%), followed by southern and island regions (0.72% and 0.67%), and northern regions (0.54%) [7,12]. Existing surveillance for viral hepatitis in Italy only monitors acute cases [8], thus limiting the system’s ability to assess the burden of both acute asymptomatic and chronic illness. Large clinical cohort studies were established to foster research initiatives, primarily on clinical grounds [12]. Epidemiological research using administrative health information and data linkage approaches have been rarely used to overcome the lack of data. A data-linkage study was carried out in Tuscany, a region with more than 3 million inhabitants, to estimate the prevalence of people with chronic HCV infection and in need of treatment by the end of 2015 [14]. This and other seroprevalence studies conducted in Italy [13], concur to an estimated prevalence of 1%.

The HCV elimination targets by 2030, including a decrease in the percentage of undiagnosed cases and an increase in treatment coverage, were made achievable by the development of direct-acting antivirals (DAA). The World Health Organization (WHO) defined numerous targets along the continuum of care for HCV to help reach this goal, including the diagnosis and treatment of 90% of persons with chronic HCV infection by 2030 [14]. The European Office of the WHO has developed a Regional Action Plan that prioritizes efforts to combat disease and adapt global targets to the regional epidemiological context [15]. The Italian National Plan for the Prevention of Viral Hepatitis (PNEV), approved by the Italian Ministry of Health in 2015 [16], included the global and regional objectives for HCV eradication.

Since 2014, patients with severe diseases in Italy have been entitled to receive free therapy at the point of care upon a specialist’s prescription. Through a monitoring registry for DAAs, the Italian Medicines Agency (AIFA) keeps track of new patients who start receiving medication (https://www.aifa.gov.it/aggiornamento-epatite-c). With the 2017 AIFA resolution, expanded access to DAA has been achieved in Italy [11] and a national registry of patients initiating treatment was established and updated regularly. At regional level, in Tuscany a triennial regional action plan was launched in 2018 with the aim of raising the treatment coverage and contributing to the elimination of HCV in the region [12]. The plan foresaw more than 6,000 HCV chronically infected individuals to be treated every year up to 2020. Yet, identification of infected individuals and linkage to care posed a challenge to the achievement of elimination targets. A national one-time screening program [17] was launched at the end of 2021 with a focus on PLP, PWUD, and the birth cohorts from 1969 to 1989, approximately corresponding to the people aged between 33 and 53.

Given these premises, the goal of this study is twofold: a) to evaluate the effectiveness of the population-level health interventions implemented since 2018 in Tuscany to expand access to HCV treatment b) to evaluate the impact of COVID-19 pandemic on the regional action plan. Interrupted time series analysis was applied on the monthly time series health administrative data to evaluate the effect of the regional plan. Study results provide valuable evidence for decision-makers including recommendations on further interventions that may be needed.

Materials and methods

Results

In this section, we mainly focus on the results obtained in the general cohort of people, in the general cohort of people aged between 33 and 53 and in the cohort that include both PLP and PWUD; however, all the other findings are available at S1 Table.

In the observed time frame, Tuscany regional health authority delivered, 561874 serological tests, 79912 PCR tests and 13098 DAA prescriptions; people in the generic population received 561874 serological tests, 79912 PCR tests and 13098 DAA prescriptions; people aged between 33 and 53 received 242725 serological tests, 25370 PCR tests and 4695 DAA, while people with exemptions (PLP and PWUD) received 15191 serological tests, 4361 PCR tests and 1210 DAA prescriptions. In the same period, 163177 people aged between 33 and 53 received a serological test to detect early signs of HCV, 10079 individuals received a screening through PCR test and 4339 were treated through DAA. S1, S2 and S3 Figs illustrate the distribution of prescribed DAA, the number of examined PCR tests, and the number of examined serological tests over time for the general population, for the general population aged 33–53 and the population that includes PLP and PWUD, respectively. As observed, the number of tests (both PCR and serological) increased over time. Similarly, the number of prescribed DAA has increased over time.

Figs 1–3 show the results obtained in the ARIMA model, in the general population, in the sub-population of individuals aged between 33 and 53 and in the sub-population of individuals that are either PLP or PWUD, respectively. Tables reporting all the estimates are available in the S1 Table. These three figures illustrate the results in the context of a counterfactual examination. In particular, the actual values of the response variable (grey dots), the predicted values of the model (blue line), the predicted values of the response variable obtained in the absence of the intervention (orange line) and the predicted values of the response variable obtained in the presence of the intervention but in the absence of the COVID pandemic (black line) are reported in these graphs. Note that the orange line and the black line represent counterfactuals (CFs) scenarios: in particular, the orange line depicts the (predicted) values of the outcome that would have hypothetically occurred in absence of the intervention and in presence of the COVID pandemic; whereas the black line depicts the (predicted) values of the outcome that would have hypothetically occurred, symmetrically, in presence of the intervention and in absence of the COVID pandemic.

Download:

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

Fig 1. The central part of the graph represents what would have happened to DAA (Direct Acting Antivirals) prescriptions for the general population if there had been no national policy on the drug’s prescriptibility (counterfactual, orange line), while the last part of the graph on the right shows how DAA prescriptions would have evolved, taking into account the negative effect of the pandemic (orange line) and in the absence of the pandemic (black line).

The blue line, on the other hand, represents the actual values measured by the registers we used. CF: counterfactual.

https://doi.org/10.1371/journal.pone.0306733.g001

Download:

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

Fig 2. The central part of the graph represents what would have happened to DAA (Direct Acting Antivirals) prescriptions for the age group born between 1969-1989 if there had been no national policy on the drug’s prescriptibility (counterfactual, orange line), while the last part of the graph on the right shows how DAA prescriptions would have evolved, taking into account the negative effect of the pandemic (orange line) and in the absence of the pandemic (black line).

The blue line, on the other hand, represents the actual values measured by the registers we used. CF: counterfactual.

https://doi.org/10.1371/journal.pone.0306733.g002

Download:

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

Fig 3. The central part of the graph represents what would have happened to DAA (Direct Acting Antivirals) prescriptions for the PWUD (People Who Used Drugs) and PLP (People Living in Prison) if there had been no national policy on the drug’s prescriptibility (counterfactual, orange line), while the last part of the graph on the right shows how DAA prescriptions would have evolved, taking into account the negative effect of the pandemic (orange line) and in the absence of the pandemic (black line).

The blue line, on the other hand, represents the actual values measured by the registers we used. CF: counterfactual.

https://doi.org/10.1371/journal.pone.0306733.g003

Finally, Fig 4 presents a comparison of the estimated ARIMA coefficients (and 95% confidence intervals) related to the effect of the intervention; S4 Fig shows the effect of the COVID pandemic for all identified sub-populations.

Download:

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

Fig 4. Comparison of the estimated ARIMA (AutoRegressive Integrated Moving Average) coefficients (and 95% confidence intervals) related to the effect of the intervention for each population group analysed.

PWUD: People Who Used Drugs; PLP: People Living in Prison.

https://doi.org/10.1371/journal.pone.0306733.g004

As can be observed, the effect of the intervention was positive and statistically significant for the general population and selected ages. However, the effect was non-significant for PLP and PWUD (Fig 4). On the contrary, COVID-19 had a negative impact on all population groups, except for PLP for whom non-significant effect was observed (figure A4 Supporting Information file).

Discussion

Here we attempt to give an interpretation to the data obtained through an interrupted time series analysis used to assess the impact of the policies of the Region of Tuscany, Italy, promulgated to achieve the goals of HCV elimination as identified by the WHO.

Firstly, it is important to highlight the effect of relaxing the criteria for access to DAA treatment on HCV diagnostic and therapeutic performance. Indeed, the 2017 AIFA resolution on extended access to DAA made access to treatment universalistic, allowing all people living with HCV to start DAA therapy regardless of the stage of liver fibrosis. This indicates that the Tuscany regional health authority raised their awareness of the importance of preventing HCV, resulting in a progressive effort to enhance screening and, consequently, the increase in initiated treatments with DAA. However, with the passage of time, after the first wave of new treatment initiations, expanding access did not result in sustained increases in the search for new cases through continuous and targeted screening, as Herink and colleagues also pointed out in their study [34]. In fact, retesting of patients already linked to care, prior to DAA treatment initiation may have contributed to the observed initial peak in screening.

A remark must be made about the high number of tests provided to women during the study period. Even before 2015, in Tuscany Region as well as in the national territory, a provision was in force - and still is - to test for HCV antibodies during the first trimester of pregnancy. This results in a high flow of tests provided to women of childbearing age, partly overlapping with the 1969–1989 birth cohort considered in the study. However, this screening effort does not correspond to a high number of HCV treatments prescribed to women, in line with other studies reporting a low prevalence of HCV among pregnant women [35].

Parallel to the increase in access to diagnostic services, which led to an increased number of new diagnoses, the number of patients initiated on treatment also saw a marked increase, particularly between 2017 and 2018. This spike in the provision of DAA treatment in the first year after the AIFA resolution can be explained by the fact that all patients already diagnosed and linked to care were recalled to start treatment. The resolution had a positive and substantial effect on the increase in the number of treatments provided because of the broadening of the inclusion criteria leading to universal access. In addition the Regional Health Authority implemented the ministerial indications and endorsed a regional HCV elimination plans, thereby favoring the linkage to care of people living with HCV infection. As the work of Kondili and colleagues shows, the effect of the national resolution on regional policies led Italy to be one of the twelve nations on course to meet the WHO’s 2030 HCV elimination targets, provided that 40 000 patients receive treatment year until 2030 [36]. This trend was also confirmed in other studies [37].

As our results show, the effect of the intervention was positive and statistically significant in the general population and the selected birth cohort. However, this effect becomes statistically non-significant in sub-populations consisting of PLP or PWUD. This finding could be explained with the fact that both PLP and PWUD are well characterized for their heightened prevalence of HCV, and they may have been targeted for HCV screening and treatment efforts before 2017. Although PWUD and PLP may be engaged in screening programs via dedicated healthcare services, namely harm reduction services or prison healthcare services, scarce evidence is available at local and national level as to the extent of such interventions in recent years [38]. On the other hand, these results could be explained by a scarce effort of public health authorities in targeting these specific populations before and after the interventions, despite robust knowledge about drug use as a risk factor, as also demonstrated in modeling studies [39].

In order to tackle this and to promote the identification of undiagnosed cases, especially among population groups where virus transmission is most frequent, a region-wide screening campaign was launched in 2023. Already since the end of 2022 in harm reduction services (SerDs) and prisons, an intensified screening effort was launched. However, it is likely that the rate of acceptance and initiation of treatment is lower in these settings, partly due to higher rate of treatment discontinuation (e.g., post-release discontinuity of care) or suboptimal compliance, and partly due to the challenges in reaching PWUDs who are not linked to harm reduction services.

In the general population, the triennial HCV elimination action plan promoted by the Tuscany Regional Health Authority promoted access to DAA and increased the coverage of HCV treatment, as demonstrated previously. This effect would have been even more pronounced if the COVID pandemic had not affected the progress starting from March 2020. The COVID pandemic had a negative and significant effect on the general population and on all examined sub-populations, resulting in a reduced number of HCV treatment initiations. This evidence has been brought to light by a number of other studies [40,41]. In particular, the work of Binka and colleagues [42], through an interrupted time series analysis, shows how public health measures related to COVID-19 had a negative effect on HCV screening and diagnosis performance, reducing the number of HCV diagnoses in Canada. In parallel and as a consequence of this, the number of treatments delivered also declined: global trends in DAA utilisation have been explored extensively in a number of studies, which concluded on a negative impact of the COVID-19 pandemic [43]. Konstantelos and colleagues have also shed light on the same phenomenon in Ontario, Canada, obtaining results overlapping with those emerging from our work in a high-income country with universalist health policies similar to Italy [44]. Because of the COVID-19 pandemic, as well as of other intervening factors, Tuscany Regional health authority failed to meet the targets for treatment set in the 2018 regional action plan, in terms both of new treatments initiated and new diagnosis.

Our research has some limitations. First of all, information on tests carried out by private providers (such as non-governmental organizations) is not included in administrative health records used to determine outpatient services and pharmaceutical services. Additionally, information on the outcomes of laboratory tests and treatments could not be retrieved. We were unable to evaluate the development of the HCV continuum of care because the laboratory test results of the patients were unavailable. Yousafzai et al. [45] documented the HCV care cascade following the introduction of DAAs in Australia in a recent study utilizing the same methodological technique by accessing HCV test results through data linkage. Second, we gathered data on treatment prescriptions; however, since data on patients’ actual drug use was not available, we used prescriptions as a stand-in for DAA uptake, even if we do not have information about the treatments’ outcomes.

In order to, link data, we identified PWUD using exemption numbers. Though not particularly for intravenous drugs, the same code is linked to all substance use disorders, which likely causes an overestimation of PWUD in our study. However, acquiring an exemption code from the SSR requires linkage to care (such as HR services), which could result in an underestimate of the real PWUD population living in Tuscany throughout the study period.

Additionally, testing services given prior to 2015 were not considered because our analysis began with the introduction of DAAs in that year. This might have caused the proportion of the Tuscan population who had undergone HCV screening at least once to be underestimated.

Conclusions

The results of this study indicate a significant upsurge in activities associated with the screening of Hepatitis C Virus (HCV) in Tuscany between 2015 and 2022. Furthermore, it is evident that the triennial action plan implemented by the regional authorities following the 2017 AIFA resolution has substantially enlarged the treatment coverage of HCV in the general population and that the effect of the policy would have been even stronger if the COVID pandemic had not affected service provision since March 2020. In the sub-population of PLPs and PWUDs the effect of the intervention was not statistically significant, suggesting that treatment coverage remained substantially stable after the implementation of the policies. To conclude, the strategic policy implemented by the Tuscany Regional Health Authority following the AIFA resolution to enhance DAAs coverage has proven to be remarkably effective in the general population, while additional efforts may be needed to target equally effectively population sub-groups such as PLPs and PWUDs. Through a targeted approach, the region has succeeded in expanding access to DAA prescriptions, demonstrating a clear commitment to combating Hepatitis C.

The findings presented in this contribution serve as compelling evidence, urging other national and subnational authorities to enhance HCV and streamline access to DAA prescriptions. In the years immediately following the AIFA resolution, the regional plan of the Tuscany Region Health Authority had an important effect in increasing the distribution of the treatment. However, it has not been as effective in actively proposing screening for population groups at high risk of transmission, nor for the general population. This is the context of the national screening program, to which Tuscany Region has adhered, that will start in full force and effect in mid-2023. The role of the screening campaign is crucial for several compelling reasons. First, timely screening enables the early detection of HCV infections, facilitating swift intervention and treatment initiation. Identifying cases in the early stages is critical to prevent disease progression and reduce the risk of severe complications. Second, promptly detecting individuals with HCV curbs the potential for further transmission within communities. Lastly, from a public health perspective, widespread screening aids in comprehending the virus’s prevalence in different communities, guiding targeted interventions and resource allocation for effective disease management.

In addition to the screening campaign, it is also crucial to facilitate access to DAA prescriptions, to enhance treatment accessibility. Simplifying the procedures for healthcare providers to prescribe these antivirals for free ensures that affected individuals, regardless of their socio-economic status or geographic location, can readily access and benefit from modern, effective treatments. From a public health perspective, early detection and treatment of HCV can lead to substantial healthcare cost savings in the long run. Indeed, by preventing the progression of the disease to advanced stages, authorities can reduce the economic burden associated with treating severe complications and improve the efficiency of the whole healthcare system.

From a broader perspective, increased screening and improved access to DAA prescriptions align with the worldwide efforts to eliminate HCV as a public health threat. National and subnational authorities play a pivotal role in contributing to international goals by implementing effective screening programs and ensuring that appropriate treatments are readily available. In conclusion, findings of this study provide evidence to support policies at national and subnational levels to booster HCV screening and simplify access to DAA prescriptions. This not only enhances individual health outcomes but also holds broader implications for public health, disease prevention, and global initiatives to eliminate HCV.

Supporting information

Acknowledgments

The authors would like to thank the Management and Healthcare Laboratory of the Sant’Anna School of Advanced Studies, in particular Professor Milena Vainieri, for facilitating access to the administrative data.