Anti-hepatitis C virus seroprevalence in the working age population in Poland , 2004 to 2014

B Walewska-Zielecka 1 2 , U Religioni 1 , G Juszczyk 1 , ZM Wawrzyniak 3 4 , A Czerw 1 , P Soszyński 2 , A Fronczak 1 1. Department of Public Health, Medical University of Warsaw, Poland 2. Medicover Sp. z o.o., Poland 3. Department of Prevention of Environmental Hazards and Allergology, Medical University of Warsaw, Poland 4. Faculty of Electronics and Information Technology, Warsaw University of Technology, Poland


Introduction
Liver cirrhosis, liver failure and hepatocellular carcinoma are possible long-term consequences of untreated hepatitis C virus (HCV) infection [1][2][3][4][5], which the World Health Organization (WHO) considers as a serious public health concern and one of the major public health priorities [6].HCV is transmitted mostly by percutaneous exposure to blood [7,8], including intravenous drug injection, which is becoming an important route, especially in developed countries [9,10].Mother-to-child transmission occurs as well; however, it is relatively uncommon, affecting an estimated 4% of children of HCV-infected mothers [11,12].
In 2005, ca 185 million people in the world, corresponding to approximately 2.8% of the global population, were estimated to be anti-HCV positive [13].The prevalence of HCV infection ranges from 1.2% to 3.8% in different parts of the world and is highest in central Asia (3.8%), east Asia (3.7%) and North Africa/Middle East (3.6%)[14,15].In the United States (US), HCV infection prevalence is at 1.6% (2.1% in men and 1.2 % in women) and higher (75% of all cases) in people born between 1945 and 1965 [16].For this reason, both the Centers for Disease Control and Prevention (CDC) as well as the American Gastroenterology Association (AGA) recommend screening for all individuals born in this period [17,18].
A study from 2014, based on comprehensive literature search anti-HCV prevalence, found the prevalence in Europe to vary from 0.9% in western Europe, through 1.3% in central Europe to 3.3% in eastern Europe [19].A report from the European Centre for Disease Control and Prevention estimates that in European Union (EU)/ European Free Trade Association (EFTA) countries over half of persons with HCV infection in 2006 are in the 25-44 year age group and overall men (64.4%) are more affected than women (35.6%) [20].From the 1990s up to 2007, new infections appeared to decline in western Europe, while they increased in eastern Europe, possibly due to rising numbers of people who inject drugs (PWIDs) in the east and effective needle sharing programmes in the west [15,21,22].
In Poland, newly diagnosed HCV infections are registered and monitored by the National Institute of Public Health since 1997 [23,24].The data are based on formal notifications from local Sanitary Inspectorates of newly diagnosed HCV infections according to the national case definition [25].A regulation of the Minister of Health of 20 September 2012 made anti-HCV tests mandatory in all pregnant women from that year onwards [26].
The latest estimates for HCV infection incidence in the country are 7.99 newly diagnosed cases per 100,000 inhabitants in 2014 [27] and, preliminarily, 11.14 newly diagnosed cases per 100,000 inhabitants in 2015 [24].HCV infection incidence is much higher in the cities (10.7/100,000 inhabitants) than in rural areas (4.82/100,000 inhabitants) and in men (8.58/100,000) than in women (7.44/100,000) [27].As acute HCV infection is usually asymptomatic, 86% of infected people in Poland are estimated to be unaware of their infection [14,28].Therefore increasing the diagnosis rate of infected persons is important [2], not only to more timely treat hepatitis C, but also to stop further spread of HCV.
Research on HCV prevalence in Poland has so far mainly focused on specific groups (healthcare workers, patients, volunteers, students, blood donors, pregnant women) or on selected areas of Poland [28][29][30][31][32][33][34][35][36].There are no epidemiological data for the prevalence of HCV in the general working age population over the whole country, especially based on a large population sample.The purpose of this study is therefore to estimate the anti-HCV seroprevalence in the working age population of Poland, using real-life data obtained from medical records of countrywide outpatient clinics, and accordingly formulate recommendations on age-related HCV infection screening.

Data source
Data were obtained in February 2015 from electronic medical records of a large countrywide outpatient clinic network operating mainly in big cities (with more than 300,000 inhabitants) representing the capitals of 11 of the 16 regions in Poland (Bialystok, Bydgoszcz, Gdańsk, Katowice, Krakow, Lublin, Łódź, Poznań, Szczecin, Warszawa, Wrocław).The clinics provide medical services predominantly to urban and suburban inhabitants with a negligible share of patients from rural areas.It is estimated that study clinics are accessible to a total of 6 million city dwellers (15% of the Polish population).

Testing for hepatitis C virus antibodies
In order to estimate the seroprevalence in the study population, only the results of anti-HCV were analysed.Anti-HCV in serum was detected by electrochemiluminescence (Roche, ECLIA) and the detection method did not change throughout the study period.All patients with positive results had been referred to special infectious disease clinical departments in order to undergo confirmatory HCV RNA tests if necessary; therefore, those results were not available in the anonymous dataset.Anti-HCV true positive results were not confirmed by immunoblotting.Such a limited approach without final confirmation of anti-HCV positivity was applied on the basis of the European Association for the Study of the Liver (EASL) recommendations, stating that immunoblotting is not recommended to distinguish false positive and true positive anti-HCV result.In order to confirm current viraemia, an HCV RNA test ought to be performed, however this was not the aim of this study [37].

Study population
The total population aimed to be investigated in the study consisted of patients who had been tested for anti-HCV at least once in the period from 2004 to 2014.The study group was extracted from the pool of all medical records of 1.5 million individuals who had been consulted by any doctor in this period.Available records included information on: unique patient number, sex, date of test, age at the date of testing, diagnosis related to the test referral and test result.Only the latest result of testing was included into the study pool, which finally comprised 61,805 single test results of unique patients.The study group was limited to working age population representatives, aged 15-64 years.The working age population was defined according to the definition of the Organisation for Economic Co-operation and Development (OECD) [39].In Poland, the working age population consists of 25 million people including 11.768 million women and 12.971 million men.

Analysis of the population tested for hepatitis C virus antibodies
The total population tested for anti-HCV was divided into 10-year age groups stratified by sex and the year of testing for time analysis.In each subgroup, the total number of patients tested was used as a denominator.

Analysis of the population testing positive for hepatitis C virus antibodies
The number of persons with a positive result for HCV antibody were available each year along with demographical data (sex and age).The rate of total positive tests was calculated and stratified by sex and age.The analyses by age group were conducted by comparing the number of patients, the number of tests and the number of positive/negative results for HCV antibody.Two classifications according to age were used.In the first classification the study population age range was split into 10 year-age groups.In the second classification, since the US data indicated a higher prevalence of HCV in people now aged 50 to 70 years [16], the percentage of positive anti-HCV test results was accordingly analysed in age groups 15 to 49 years and over 50 years.
The mean rates of positive patients were analysed over time by regression analysis and stratified by sex.

Analysis of testing and positive tests by referral group
A number of referrals for anti-HCV test (n = 36,356) had preliminary diagnosis information (according to ICD-10 coding) [40].We compiled those diagnoses into specific groups for further analyses (Table 1).Both 3-digital and 5-digital ICD-10 codes were aggregated.

Statistical methods
Data were analysed using STATISTICA (data analysis software system), version 12, (www.statsoft.com)StatSoft, Inc. (2014) US, to calculate the incidence of newly diagnosed cases per year, and the prevalence in the entire examined population.The independent-sample t-test was used for normally distributed variables, and the nonparametric Mann-Whitney U test was used for not normally distributed parameters.Significance was set at p < 0.05.Using linear regression analysis, the trend of the number of the incidence as a function of time (years) was calculated and the R-square value evaluated the goodness of fit of the regression.

Overall characteristics
A total of 61,805 single patient records were considered in the study, spanning the period from 2004 to 2014 (Table 2).Men (n = 19,531) accounted for 31.6% of the total study group.The overall average age of patients was 34.4 years (standard deviation (SD): 8.6).The average age of men was 36.5 years (SD: 9.6).The average age of women was 33.4 years (SD: 7.9) (Table 3).

Analysis by age group
The most represented age group in terms of number of individuals was the one comprising 25 to 34 yearolds (n = 35,047 patients; 56.7%) and the least numerous group comprised persons over 55 years (n = 2,626 patients; 4.2%) (Table 4).

Time analysis of testing practices
The number of patients examined for anti-HCV increased steadily with time, from 815 patients in 2004, to 14,963 in 2014 (Table 2).
The percentage of all medical-facility-patients tested yearly increased from 0.9% (815/88,177) in 2004 to 4.0% (14,963/376,637) in 2014.Data showed a growing proportion of women being examined.In 2004, a similar number of men and women underwent anti-HCV tests (50.3% (410/815) of women and 49.7% (405/815) of men), whereas in 2014, women accounted for 79.1% (11,620/14,693).This increase may reflect legal requirements for prenatal care during pregnancy in Poland, with HCV testing becoming compulsory for pregnant women from 2012 onwards (Table 5) [26].

Time analysis of patients testing positive for hepatitis C virus antibodies
An analysis of the data in the years 2004 to 2014 suggests a downward trend for the proportion of positive anti-HCV results (mean positive  anti-HCV = -0.0017× year + 3.3715; R2 = 0.7558) (Figure ).

Referral group testing and test results
We analysed the diagnoses ascribed to each anti-HCV test referral.The predominant reason for anti-HCV testing was pregnancy and pregnancy-related conditions -44.4% (16, 6).

Discussion
Our study presents an evaluation of anti-HCV prevalence in a large country-wide sample of (sub)urban working-age Polish people between 2004 and 2014.Data from electronic medical records of ambulatory patients visiting doctors due to different conditions, including prophylactic and screening reasons, were analysed.The overall anti-HCV prevalence in our study was 1.5%.Similar to previous studies [29,30] and studies from other countries [41][42][43], anti-HCV positivity was significantly more frequent in men than women (2.0% vs 1.3% respectively; p = 0.0001).
We also found that in contrast to younger age groups (15 to 49 years), anti-HCV prevalence in people aged between 50 and 64 years was higher (2.7% vs 1.4%; p < 0.0001), and surprisingly more frequent in women (3.0%) than men (2.4%) although, this difference was not statistically significant.Higher HCV infection prevalence in people born before 1965 has also been observed in the US [16], therefore, the recommendation of CDC and AGA to screen people born before 1965 [17,18] might also be justified in Poland and could be implemented as part of primary healthcare.According to the authors' own research on 16,130 pregnant women, the prevalence of HCV in this group was only 0.8% which is close to the European average (1%) [44].Low anti-HCV prevalence in pregnant women and high prevalence in people of post-reproductive age might be a subject of debate in terms of allocating effective financial resources for HCV screening in these two groups.
To our knowledge, our study constitutes one of the currently largest performed in the Polish population of working age.Indeed, although previous studies in the country have attempted to assess HCV prevalence, these have either been conducted either some time ago, or have been mainly based on small samples and/ or on selected population groups -pregnant women, students, blood donors or deceased organ donors -with, in some cases, only a single district or town in the country considered [28,29,[31][32][33][34][35]45].For example, the study by Bielawski et [47].In our study, we had no access to individual medical records; therefore, intravenous drug use could not be accounted for.Moreover, we did not have information on patients' profession either, so we could not evaluate any possible occupational risk.
Second, our study only analysed HCV antibody prevalence, and western blot tests were not performed.Thus results do not distinguish between current infections and probable infections in the past (resolved infection).
A final diagnosis of current HCV infection requires the finding of HCV RNA in serum samples by RT-PCR.We had no access to HCV RNA results that were stored in the form of scans and required patients' consent for access.The results of this study can therefore not be compared with any studies using the EU HCV case definition [48].A Polish study in 2011 showed that only 31% of those with HCV antibodies were also positive for HCV RNA by RT-PCR [30].Moreover the sensitivity of the electrochemiluminescence (Roche, ECLIA) assay used in this study is very high, so specificity will be lower, which may result in false positive results.Taking these points into consideration, the prevalence of current HCV infection in the Polish urban working population is likely to be lower than 1.5%.Further studies on positive anti-HCV test results and HCV RNA detection may reveal if HCV infection is resolved more frequently than has been presumed up to now (ca 30% of cases being resolved [49]).
Finally, although our study was conducted on a large number of patients, another important limitation was the inclusion of only people living in big cities and their suburbs.One risk factor for HCV infection is the use of medical care, which is less frequent in inhabitants of rural areas.Accordingly, anti-HCV prevalence in rural areas has been shown to be lower than in urban areas [27,50].Therefore our results cannot be extrapolated to the whole population and it can be assumed that the prevalence of HCV infection among people of working age in the country may be lower than 1.5%.

Conclusions
There is evidence that an improvement of diagnostics and treatment effectiveness may significantly reduce the burden of HCV infections in Poland [5,51].A study using a modelling approach estimated that, until 2030, the HCV prevalence is projected to decrease by 5%.In contrast, an increase in the number of treated patients to 15,000 yearly would reduce the number of total infections by 90% until 2030, which would also contribute to a decrease of HCV-mortality by 80% [49].The results obtained in this study suggest that the proportion of people infected with HCV in Poland in the working population is decreasing, which may be a consequence of increasing social awareness, including preventative activities after or before exposure to blood-borne infections.Moreover, a higher prevalence of anti-HCV was found in the population of post-reproductive age.We therefore recommend screening HCV tests mainly in individuals over 45 years-old.Examining healthy and young people should not be carried out as part of screening, however testing may be recommended to individuals who are subjected to risk factors.The continuous monitoring of HCV prevalence and incidence in Poland is important to estimate the resources needed for screening and treatment as well as their costs.Knowing the age groups at higher risk for infection will help to establish recommendations for more effective detection of cases of HCV infection, which in turn is also crucial to reduce further transmission.

Figure 1 F
Figure 1 Proportions of hepatitis C virus antibody positive tests among the study population, stratified by year and sex, Poland, 2004-2014 (n = 61,805 patients)

Table 2
Annual numbers of patients tested for hepatitis C virus (HCV) antibodies and proportions testing positive, stratified by sex, Poland, 2004-2014 (n = 61,805 patients) (3sitive patients was 36.8 years (SD: 9.8).The average age of anti-HCV positive women was 36.0 years (SD: 9.8), and the average age of men with positive test results was 37.8 years (SD: 9.7).4%;132/9,444), and highest for women aged over 55 years(3.2%;43/1,394).For the youngest group comprising 15 to 24 year-olds, the value 1.5% (33/2,234) was similar to that of the group of 35 to 44 year-olds (1.4%; 132/9,444).In the group of tested men, the smallest proportion of infections was found in the 15 to 24 years age group (1.6%; 19/1,177) and increased

Table 3
Characteristics of the study population and that testing positive for hepatitis C virus antibodies, Poland, 2004-2014 (n = 61,805 patients) [16]use a higher prevalence of HCV was reported in 50 to 70 year-olds in the US[16], the percentage of positive anti-HCV test results was also analysed in age groups 15 to 49 years (representing 92.1% of those tested 56,921/61,805) and over 50 years (7.9%; 4,884/61,805).A higher percentage of anti-HCV positive patients was found in those aged over 50 years

Table 4
Results of anti-hepatitis C virus tests stratified by patient age groups and sex in a study estimating hepatitis C seroprevalence, Poland, 2004-2014 (n = 61,805 patients) HCV: hepatitis C virus; SD: standard deviation.

Table 6
Primary diagnoses resulting in the referral for anti-hepatitis C virus tests, Poland, 2004-2014 (n = 36,356 patients) Excluding people with elevated transaminase level as a reason of additional anti-HCV test.
b c Most often before surgical procedures.
al. in 1999, which is still the point of reference for many epidemiological studies on HCV infection in Poland, was conducted on a group of 2,561 volunteers enrolled by a laboratory in Gdansk in response to press advertisements.