A time-series analysis of morbidity and mortality of viral hepatitis in Venezuela, 1990–2016

This is a time series study involving records of morbidity and mortality by VH in Venezuela reported during the period from 1990 to 2016. The Venezuelan population was taken as the denominator of the morbidity and mortality rates, according to the Venezuelan National Institute of Statistics and the 2016 population projections from the latest census published on the website of the responsible Venezuelan agency [16].

The setting for this study was Venezuela with its 25 states. The country has a total geographical area of 916,445 km², located in the northern part of South America, with a 2016 projected population of approximately 31,028,637 inhabitants [16]. Venezuela is characterised by significant regional inequalities that have been accentuated by the recent political, social, and economic crisis [17] and have precipitated the emergence and re-emergence of vector-borne and immuno-preventable diseases [11, 18].

The structure, design, and operation of the Venezuelan health system is an official agency whose objective is to monitor morbidity and mortality occurred in the national territory through the “Ministerio del Poder Popular para la Salud” (Ministry of Health).

In relation to morbidity, there are the Mandatory Notifiable Diseases (infectious and non-infectious diseases) that are classified according to their periodicity of reporting in immediate, daily, or weekly notification (in the case of hepatitis, the report is weekly) [19]. Notification, denunciation, and reporting are routinely and mandatorily performed by law and depend on the physicians of the medical care units, discriminated into urban outpatient clinics, rural outpatient clinics, and all types of hospitals including specialised ones. The information obtained by these units must be systematically and thoroughly collected by local or regional epidemiologists and subsequently used to generate the Venezuelan Epidemiological Bulletins.

Furthermore, death statistics come from the death certificates registered continuously in the municipalities, civil registry units, prefectures, and civil headquarters throughout a calendar year. The country’s information sources report each month the deaths registered in the previous month, based on the death certificates, which are filled out the first five days of each month by officials of the National Institute of Statistics and subsequently published in the mortality yearbooks. The quality of death registration data obtained from the Venezuelan epidemiological surveillance system has been previously rated as high [20].

In this study, data were extracted from Epidemiological Bulletins (not published since 2016) and official yearbooks published from 1990 to 1995 in printed physical version and from 1996 to 2016 in digital version from the portal page of the responsible Venezuelan agency [21]. No population was excluded. All ages and all sexes were included. The specific diagnoses of VH according to duration, acute or chronic, or according to viral type (A, B, C, D or E) are clinical, epidemiological, or laboratory-based. Unspecific viral hepatitis (UVH) are all those VH where the diagnosis does not specify the viral type either by clinical diagnosis, presumptive epidemiological diagnosis, or confirmation by paraclinical tests. This includes VH without further specification, as listed in the WHO’s International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) [22].

The Venezuelan mortality yearbooks use the coding of the ICD-10 since 1996: VH (B15–B19) and sequelae of VH (B94.2), referred to for the purpose of this paper as total viral hepatitis (TVH) (B15: acute hepatitis type A; B16: acute hepatitis type B; B17: other acute viral hepatitis; B18: chronic viral hepatitis; B19: unspecific viral hepatitis) [22]. The coder from the mortality statistics department analysed the certificates according to the causes of death recorded (direct cause, antecedent causes, and basic cause) and the approximate time interval between the onset of the disease and death. Mortality due to HCV infection and sequelae of VH were incorporated into the yearbooks in 1996, while cases of VHC were incorporated into the epidemiological bulletins from 2005 onwards. Moreover, in 2004–2005 there was a change in coding in the statistical section of the Ministry of Health (where ICD-10 coding is assigned, in terms of the time of VH and sequelae of VH according to the basic cause of VH).

Statistical analysis was performed using Statistical Package for the Social Sciences version 21 (International Business Machines Corporation, Armonk, NY, USA) and figures was generated using Microsoft® Excel® version 2019 (Microsoft, Redmond, WA, USA). For the calculation of rates, the nominator was the specific statistics from Epidemiological Bulletins and official yearbooks in absolute numbers per year, and the denominator was the specific population according to official data from the Venezuelan National Institute of Statistics, multiplied by a constant of 100,000 inhabitants.

The variables were analysed with absolute and relative values, calculation of year-specific rates, averages of lapses, and indices. For the calculation of average rates, triennia were used from the first three chronological years of the series in ascending order. The percentage rate was calculated based on 100% of the first data in the time series in consecutive order. The annual case fatality rate was calculated as a division between annual mortality and morbidity and is expressed as a percentage.

For the study of variable correlations, Pearson’s correlation coefficient was used with a confidence level of 95% and 99%, α error = 0.05 and 0.01, respectively. R value > 0.8 was considered very strong; 0.6 to 0.79, strong; 0.4 to 0.59, moderate; 0.2 to 0.39, weak; and < 0.2, absent. The polynomial trend line for VHB and sequelae of VH mortality was plotted, and the formula equation and coefficient of determination (R²) were calculated.

VHA infection is closely associated with unsanitary food and water consumption, poor sanitation, and poor personal hygiene [4]. In this study, epidemic behaviour of VHA and UVH match with the large troughs and landslides caused by torrential rains that occurred during those years in several regions and localities of the country. The activity of tropical storm Bret in 1993 caused floods and material losses through Venezuelan territory, including Monagas, Sucre, Nueva Esparta, Distrito Federal, Aragua, and Carabobo states, among others, and it is estimated that at least 120 people died, several and more than 4,000 victims as a result of this tropical storm [25]. Other catastrophes such as the one that occurred in Vargas state in 1999 and 2005, which destroyed housing and sanitation infrastructure, including the collapse of sewage and water systems, mainly in Vargas, Miranda, and Falcón states [9]. Similarly, in 2005, in the Mocotíes tragedy in Mérida state, land communications were interrupted, among other natural disasters caused by rain in Aragua and Miranda states, with deaths and great destruction of populations such as Araira, with many refugees [10]. The vaccine has had a significant effect in reducing HAV infections, although coverage rates remain lower compared to other childhood vaccines [26]. Countries in the region, such as Argentina, have had successful experiences following the incorporation of the vaccine into the national immunisation schedule, with rates declining sharply from 113.3/100,000 in 2004 to 1.4/100,000 in 2011. Throughout this period, vaccination coverage was over 90% nationwide [27]. In contrast, in Venezuela, in the case of hepatitis A vaccination, it is not included in the Venezuelan Expanded Programme on Immunisation, it is only applied in the private sub-sector and in the years of better economic boom, prior to 2013, it reached between 10% and 15% of vaccination coverage, according to personal and press reports. However, there is no reliable data regarding private sector indicators on vaccination [28]. Finally, the WHO Global Health Observatory [29] shows the most relevant indicators for monitoring VH, such as VHB and VHC, from 2000 to 2020. However, it does not keep records of VHA, morbidity data or VH seroprevalence, which further limits understanding of what is happening in Venezuela.

The spectrum of acute HBV infection ranges from asymptomatic infection to self-limited hepatitis, to fulminant hepatitis. A third of acute infections in adults are symptomatic, and fulminant hepatitis occurs in less than 1% of cases, with a mortality of about 70% [30]. Chronic HBV infection occurs in approximately 5% of those infected in adulthood, compared to 90% of those infected in early childhood [31, 32]. Therefore, the global incidence of chronic HBV infection is largely due to vertical, mother-to-child, and early childhood transmission, and this is the focus of the WHO elimination plan. This study found a negative correlation between VHB and VH mortality and chronic sequelae of VH; that is, as VHB deaths decrease, deaths from sequelae increase; this correlation may be explained by the way deaths from VHB were reported before 2004, which were probably deaths associated with sequelae of VH. The disruptive drop in VHB deaths and the rise in deaths from sequelae of VH evidenced in 2005 may be explained by the change in coding that begins to consider the basic cause of death. This is supported by the analysis of the mortality yearbooks, where the majority of deaths (90%) by malignant tumour of the liver and intrahepatic bile ducts, non-alcoholic liver fibrosis, and cirrhosis occur in adults older than 44 years, coinciding with the most frequent age group (older than 35 years) where the majority of deaths due to sequelae of VH are found (>96%), suggesting that they are a result of chronic hepatitis.

In 2000, the vaccination schedule against hepatitis B in children under one year of age was intensified, and, in 2001, given the high incidence of HBV infection, a vaccination strategy was launched in new-borns to prevent vertical transmission of this disease. Over the years, coverage has progressively increased, reaching 88% in 2006, which may explain the decrease in both morbidity and mortality due to HBV in the last years of this study period [23]. Therefore, the primary intervention to HBV is the vaccination. However, vaccination coverage rates have been declining in recent years [11], including hepatitis B vaccination, reaching a coverage of 56% in 2021 [23]. Although there was a sudden decrease in VHB mortality and an increase in sequelae of VH since the time of the coding change, we consider that the maintenance of this trend may also be due to the impact of vaccination, given the decrease in deaths in children in correlation with the increase in vaccination coverage. Moreover, we argue that deaths due to sequelae of VH will continue to increase at the expense of the population that did not have access to HBV vaccination and who are currently chronically infected.

Conversely, the majority of cases of acute HCV infection occurs without clinically overt disease, but leads to a chronic infection in around 75–85% of people; over the course of 20–30 years, a proportion of patients will progress to liver cirrhosis and hepatocellular carcinoma [33]. The prevalence of HCV infection varies according to geographical area and different risk groups [34]. In Venezuela, there are few studies aimed at determining the frequency of HCV infection in the general population. However, prevalence of HCV infection has been reported from 0.6% to 2.1% [34, 35]. Most seroprevalence studies in the country have focused mainly on small regions (mainly the city of Maracaibo in the northwest of the country), Amerindians and high-risk groups such as sex workers, haemodialysis patients, and prisoners, with prevalence ranging from 0% to 71% [14, 36‐44], with the expected trend of increasing prevalence of HCV infection exposure as a function of age. Similar to VHB, VHC mortality correlated significantly with death from sequelae, suggesting changes in epidemiological reporting from 2004 onwards. In addition, studies in Latin America and the Caribbean have reported several other risk factors for HCV infection and eventual hepatocellular carcinoma in these high-risk groups [45‐47]. HCV was the aetiology in one third of 116 Venezuelan patients diagnosed with hepatocarcinoma [48]. However, information on HCV-related sequelae in Venezuela remains very limited.

The limitations of this study are several: (1) the unavailability of bulletin publications after 2016 prevent in country morbidity and mortality analysis from being more up-to-date; (2) data quality limits the certainty of our results, although the bulletins are guided by the ICD-10, reporting is done by professionals from all over the country, in different socio-economic contexts, which probably limited the molecular confirmation of VH; (3) uncertainty about the size of the populations, the denominators of rates that are extrapolated from a census of several years’ duration and during a political and economic crisis of great dimensions that has generated significant migration of Venezuelans, many of them of productive age, with an important impact on the size of the general population. It is estimated that there has been a slow migration since 2000, which intensified in the last 10 years [49]. Finally, the information available on VHD and VHE in the bulletins is null and void, which is why they could not be analysed in this time series. Despite these limitations, the trends in morbidity and mortality behaviour provided by the Venezuela’s Ministry of Health offer a better understanding of the behaviour of VH in the country to guide effective responses to this important public health problem.