Hepatitis C Prevalence: Our Study Method

How Medivo Will Estimate Hepatitis C Prevalence

By Carol Smyth, MD & Brian Saindon


If you haven’t been following us so far, this is our third blog in a series that explores Hepatitis C prevalence.  Our experts are conducting an analysis of HCV+ patients within Medivo’s lab test database, which will identify patients that have been diagnosed, as well as those that may not have been.  Our goal is to enable healthcare professionals to identify where treatment is needed and implement education programs to increase testing and improve patient health. Last week we addressed how to improve HCV surveillance to better estimate HCV prevalence.  If you missed it, check it out here.

In this blog post, we will examine available data which reports the overall and age-stratified counts of patients diagnosed with Hepatitis C (HCV) in New York State.  We will use two primary data sources to report on HCV counts in New York State: New York State Department of Health (NYS DOH) and The New York City Department of Health and Mental Hygiene (NYC DHMH). We will also discuss Medivo’s plan to identify similar metrics using our lab database so that we can compare our counts to theirs in order to get a sense of whether our HCV counts differ from those being reported by health officials.

The latest data from the New York State Department of Health shows 7,562 patients as being diagnosed with “HCV past or present” as of July 15, 2015 in New York State, not including New York City. This number reflects 2014 data reported to the NYS DOH as of July 2015.

The New York City Department of Health and Mental Hygiene released the “Hepatitis B and C Annual Report” in January 2016 which identifies 93,579 individuals as being diagnosed with chronic HCV during the 2011 through 2014 time  Of this group of people, 7,691 were identified as newly reported in 2014.  
Medivo experts will use Medivo’s laboratory database in order to generate counts for HCV diagnosed patients within the same regions and time periods those reported by the NYS DOH and NYC DHMH (see Chart 1).  We will compare the Medivo metrics to those identified in Chart 1 in order to identify whether the NYS DOH and NYC DHMH captures the entire HCV+ patient population in New York State.


Chart 1: Patients Diagnosed with HCV in 2014 as reported by New York City Department of Health and Mental Hygiene  & NYS Department of Health 

Hepatitis C prevalence: DOH DHMH

This graphic identifies that New York City, as reported by the NYC DHMH, has 129 more individuals identified as HCV+ than the count of HCV patients in New York State (excluding NYC) as reported by the NY DOH.   

In addition to the overall counts, both data sources provide age stratification breakdowns of these overall HCV patient counts.  See Chart 2 for the number reported by these sources stratified by age group.  

Chart 2: Patients Diagnosed with HCV Stratified by Age Group as reported by New York City Department of Health and Mental Hygiene  & NYS Department of Health  


Given the difference in overall counts, it is challenging to visually compare the age group distribution between the two sources.  In order to identify differences in age distribution of HCV patients between the two sources, let’s take a look at the percent distribution of age groups in Chart 3…


Chart 3: Patients Diagnosed with HCV Stratified by Age Group: Percent Distribution for New York State Excluding NYC*


Literature recommends targeting HCV interventions according to prevalence and risk. However, in 2012, the Centers for Disease Control and Prevention issued a call to action to test all Baby Boomers (born between 1945 and 1964) for HCV, based on a high risk of undiagnosed infection in this group.  Identifying differences in age distribution can help improve targeting HCV surveillance and interventions.  You will notice that, in both locations (within and outside of NYC) the majority of patients with HCV fall into the 50-59 or 60+ age groups., as per the CDC recommendations to test the Baby Boomer age group based on year of birth and not on risk factors.

You will notice in Chart 3, that one difference between locations to note is the 20-29 age group. The NYS DOH data shows that this age group accounts for 19.4% among all HCV+ patients that are located in outside of the city whereas the NYC DHMH reports this age group to account for only 9.7% of all HCV+ patients living in NYC.  This data suggests targeting interventions towards the 20-29 age group in New York State (outside of NYC) may be needed.


Study Introduction

As part of our research initiative, we will use the Medivo laboratory database in order to generate similar counts of patients diagnosed with HCV in New York State.  We will compare our HCV counts with the data that is currently available from the NYS DOH and the NYC DHMH as presented in Tables 1 – 3.

Medivo data relevant to this study include specific laboratory tests used to diagnose HCV such as the HCV antibody test, HCV genotype tests and viral load tests.  We will apply Medivo diagnostic logic to these test results in order to identify individuals who test positive for HCV.


Data, Study Design, & Analytic Plan

Data for this study include all HCV antibody tests, HCV Genotype tests and Viral Load test results from Medivo’s nationwide lab test database during the January 2011 through May 2016 time period.  Only laboratory tests from New York patients will be included in this study.  

We will explore simple descriptive statistics among patients diagnosed with HCV.  Data will include patients that are identified as diagnosed with HCV between January 2011 through May 2016.  See the HCV Diagnostic Criteria section below for details on our classification of a patient diagnosed with HCV.

Medivo identifies HCV diagnostic criteria similar to the CDC’s laboratory diagnostic criteria for HCV, Past/Present.  Medivo diagnostic logic will identify a patient as being diagnosed with HCV  if s/he meets the following criteria:

  • has an HCV antibody test greater than 8 or
  • has an HCV antibody test labeled as “reactive” or
  • has any test for HCV genotype or
  • has any test for viral load quant or
  • has any test for viral load qual

Descriptive statistics will include counts of patients diagnosed with HCV in two groups: those who live in New York State, exclusive of New York City and those who live in New York City.  Both groups will be stratified by age group.  We will compare overall and age-stratified counts for New York State excluding NYC to counts available by the New York State DOH for 2014. We will compare Medivo counts for New York City to the counts released in January 2016 by the New York City Department of Health and Mental Hygiene.  



Limitations of this study include sampling bias, since the Medivo laboratory database is a non-random sample of the US population.  This analysis will under represent homeless or incarcerated individuals.  Furthermore, this study will not capture patients who are not seeking treatment/being tested (asymptomatic).  This analysis also does not account for population shift so it is possible that individuals with HCV may have moved out of New York State since the beginning of the reporting period.   

Lastly, we are limited in comparing our HCV counts to those from the NYS DOH and the NYC DHMH due to differences in HCV case definition as discussed in blog 2.  



Databricks and Spark software had been used for data processing and statistical computations.  All data had been stored using Amazon Web Services.  Charts and visualizations in this blog and the next blog are created using Tableau Software v 9.1.



It has long been suspected that Hepatitis C prevalence has been severely underreported, but due to lack of attention and funding, little has been done to prove or rectify these circumstances.  If Medivo’s study suggests that HCV is much more prevalent than health organizations have reported, our goal is to spark positive change in the testing, surveillance, reporting, and treatment of HCV.  If we in the medical community can identify more patients, we can get them the right treatment at the right time, improving and even saving lives.



This is the third blog post in a five-part series. Our next blog post will include the results of the analysis outlined in this blog post.  We welcome your comments and feedback regarding any part of this blog series. Please comment below, or reach out to us at info@medivo.com.


  1. https://www.health.ny.gov/statistics/diseases/communicable/2014/
  2. http://www1.nyc.gov/assets/doh/downloads/pdf/cd/hepatitis-b-and-c-annual-report.pdf
  3. Aspinall, EJ, Doyle, JS, Corson, S et al. (2015). Targeted hepatitis C antibody testing interventions: a systematic review and meta-analysis. European Journal of Epidemiology, 30(2), 115-129.

Previous Post

©2016 Medivo Inc., All Rights Reserved.

Up Next

Terms of Use | Privacy | © 2018 Prognos. All rights reserved