| |
(premature) Deaths among People with Hepatitis C in New York City, 2000 - 2011
|
| |
| |
Download the PDF here
HCV+ die at avg age of 60, co-infected at 52 vs 78 for those with neither disease...the same trend is seen in this study for age at death due to cancers, heart disease, diabetes
Clinical Infectious Diseases Advance Access published February 12, 2014
Jessie Pinchoff1,2, Ann Drobnik1, Katherine Bornschlegel1, Sarah Braunstein1, Christine Chan1, Jay K Varma1, Jennifer Fuld1
1New York City Department of Health and Mental Hygiene, Queens, NY, USA 11101
2Johns Hopkins Bloomberg School of Public Health, Department of International Health, Baltimore MD, USA 21205
"HCV-infected adults are at increased risk of premature death compared with persons without HCV infection, and that the HIV co-infected are at exceptionally high risk of premature death....earlier diagnosis is crucial......Though HIV co-infected persons were reported with HCV at a younger age, a greater proportion of those who were mono-infected died within three years of HCV report. The large proportion of deaths within three years in the HCV mono-infected group may indicate that, for many, testing for HCV occurs only when they present with symptoms, as HCV is often asymptomatic.....To decrease mortality for people with HCV and decrease the proportion of people with HCV who die within three years of diagnosis, improvements in testing, care and treatment are needed.....With antiviral therapy, recent improvements in HCV treatment and more expected in the coming years, it is possible to cure most patients of HCV infection; cure is associated with decreases in all-cause and hepatitis-related mortality"
"Persons with HCV mono-infection died at a significantly younger age (median 60.0 years) than those with neither infection (median 78.0 years) (Table 2). Persons with HCV/HIV co-infection died at a significantly younger age (median 52.0 years) than the HCV mono-infected and those with neither infection......The adjusted odds of death for HCV mono-infected and HCV/HIV co-infected decedents were lower for cardiovascular-related causes, non-liver cancers, and diabetes/obesity in comparison to those with neither infection (p<.05 for all comparisons) (Table 4); however, this is because these two groups were more likely to die at younger ages of other causesÓ
from Jules: in Table 1 you can see among HIV/HCV co-infected death rates are 3 times higher among blacks & latinos vs whites, although this is not seen among mono-infected, this might be because co-infection rates are much higher among black and latinos due to higher rates of IDU vs HIV+ MSM. In table 5 below you can see HCV mono and co-infected, particularly co-infected die at much earlier ages due to cancers, heart disease and diabetes.
Deaths among People with Hepatitis C in New York City, 2000 - 2011
Abstract
Background. Infection with hepatitis C virus (HCV) increases the risk of death from liver and non-liver related diseases. Co-infection with HIV further increases this risk.
Methods. Surveillance data (2000-2010) and mortality data (2000 - 2011) maintained by the New York City Department of Health and Mental Hygiene (DOHMH) were deterministically cross-matched. Factors associated with and causes of death among HCV-infected adult decedents were analyzed.
Results. Between 2000 and 2011, 13,307 HCV mono-infected adults died, and 5,475 adults co-infected with HCV/HIV died. Decedents with HCV mono-infection were more likely to have died of liver cancer (OR=9.2), drug-related causes (OR=4.3), and cirrhosis (OR=3.7) as compared with persons with neither infection. HCV/HIV co-infected decedents were more likely to have died of liver cancer (OR=2.2) and drug-related causes (OR=3.1) as compared with persons with neither infection. Among co-infected decedents, 53.6% of deaths were attributed to HIV/AIDS; and 94% of deaths occurred prematurely, before age 65. Among persons with HCV who died, over half died within three years of a hepatitis C report to DOHMH.
Conclusion. HCV-infected adults were at increased risk of dying and of dying prematurely, particularly from conditions associated with HCV, such as HIV/AIDS or drug use. The short interval between HCV report and death suggests a need for earlier testing and improved treatment.
Results
Age at death
Between 2000 and 2010, 128,444 people were reported with HCV in NYC, 18,291 (14%) of whom had also been reported with HIV. Of 110,153 persons with HCV mono-infection, 13,307 died between 2000 and 2011, and 8,525 (64.1%) of these died prematurely (Table 1, Table 2). Of those HCV/HIV co-infected, 5,475 died between 2000 and 2011 and 5,146 (94.0%) were premature. In contrast, of 619,254 adult deaths recorded in NYC with neither disease, 156,499 (25.3%) were premature. Over half of decedents died within three years of HCV report; the difference between the mono-infected and co-infected groups was not statistically significant (Table 2).
Persons with HCV mono-infection died at a significantly younger age (median 60.0 years) than those with neither infection (median 78.0 years) (Table 2). Persons with HCV/HIV co-infection died at a significantly younger age (median 52.0 years) than the HCV mono-infected and those with neither infection. HCV/HIV co-infected persons were reported with HCV at a significantly younger age (median 48.8 years) than the HCV mono-infected (median 56.9 years).
Causes of Death
Of the disease categories examined, persons with neither infection were most likely to die of cardiovascular causes (46.6%) and non-liver cancer (23.7%) (Table 3). Decedents with HCV mono-infection died from cardiovascular causes (26.3%), followed by non-liver cancer (16.1%), hepatitis C (11.6%), liver cancer (8.7%) and drug-related causes (7.6%). The greatest proportion of persons HCV/HIV co-infected died from HIV/AIDS related causes (53.6%) (Table 3). Bivariate analysis indicated that both the HCV mono-infected and co-infected were significantly more likely to die of HCV, liver cancer, cirrhosis, and drug-related causes, and less likely to die of cardiovascular diseases and non-liver cancers than those with neither infection (Table 3).
The adjusted odds of death for HCV mono-infected and HCV/HIV co-infected decedents were lower for cardiovascular-related causes, non-liver cancers, and diabetes/obesity in comparison to those with neither infection (p<.05 for all comparisons) (Table 4); however, this is because these two groups were more likely to die at younger ages of other causes (Figures 1- 3). HCV mono-infected persons had significantly higher odds of dying of liver cancer (OR=9.2), drug-related causes (OR=4.3), and cirrhosis (OR=3.7) as compared with persons with neither infection (p<.05 for all comparisons) (Table 4). HCV/HIV co-infected persons had significantly higher odds of dying of liver cancer (OR=2.2) and drug-related causes (OR=3.1) (p<0.05 for all comparisons) than decedents with neither infection. The magnitude of the associations between death due to liver cancer or drug-related causes and HCV mono-infection was higher than the magnitude of the associations between these causes of death and HCV/HIV co-infection.
The age patterns of HCV-related deaths for these eight leading causes followed different trends for persons without HCV or HIV, those with HCV only, and those co-infected with HCV/HIV. Among adults with neither infection, non-liver cancers and cardiovascular causes were the leading causes of death in all age categories; drug-related causes accounted for 16.6% of deaths in the youngest quartile but declined with increasing age (Figure 1).
Among HCV mono infected adults, drug-related causes were the leading cause of death in the youngest quartile, representing almost one-third of deaths (32.9%) (Figure 2). In the two middle age quartiles HCV-related causes (comprised of HCV, liver cancer and cirrhosis) were responsible for over one-third of deaths, surpassing all other causes. Cardiovascular causes were responsible for the greatest proportion of deaths in the oldest quartile (45.7%). The median ages at death for those reported with HCV mono-infection was lower as compared with those with neither disease for all causes, with the exception of drug-related causes (Table 5).
Among persons with reported HCV/HIV co-infection, HIV/AIDS related causes were responsible for the greatest proportion of deaths in all age categories, declining from 65.4% in the youngest quartile to 41.2% in the oldest quartile (Figure 3). Drug-related causes accounted for 14.2% of the youngest age quartile, declining with age. Cardiovascular causes, non-liver cancers and liver-related causes increased with age, but all represented less than one-third of deaths. The median ages at death for those reported with HCV/HIV co-infection were lower as compared with those with neither disease for all causes, with the exception of drug-related causes and HIV/AIDS (Table 5).
Discussion
These findings suggest that there is much work to be done to improve outcomes for people with HCV in NYC. The findings are consistent with prior studies showing that HCV-infected adults are at increased risk of premature death compared with persons without HCV infection, and that the HIV co-infected are at exceptionally high risk of premature death [4-7]. Though HIV co-infected persons were reported with HCV at a younger age, a greater proportion of those who were mono-infected died within three years of HCV report. The large proportion of deaths within three years in the HCV mono-infected group may indicate that, for many, testing for HCV occurs only when they present with symptoms, as HCV is often asymptomatic until significant liver damage has occurred [4, 14]. This is supported by studies that have found that between 28-75% of persons with HCV in the US do not know they are infected, and many of those who do are not receiving optimal medical management [23, 24].
The leading causes of death overall in both the mono-infected and co-infected groups were extra-hepatic, as previous studies have found [4-6]. Given that advanced liver disease can cause a wide range of systemic problems [25], deaths from extra-hepatic causes may also be related to HCV infection. Among co-infected persons, HCV may play a role in death even when not listed as the underlying cause, as national coding guidelines specify that a person with HIV be listed as dying from HIV/AIDS, regardless of other HCV or chronic infections being present. Co-infected persons died from liver-related causes at younger ages than persons with HCV mono-infection, supporting previous evidence of faster progression of liver disease in persons with HIV [16, 17].
Deaths from drug-related causes were highest in the youngest age quartile, surpassing all other causes of death for the HCV mono-infected group, as has been noted in other studies. [5] HCV treatment programs should incorporate overdose prevention and drug treatment, including buprenorphine prescription, into their programs to prevent premature deaths [26]. Similarly, harm reduction and drug treatment service providers are uniquely positioned to provide services that prevent new HCV infections. This finding also reinforces the need for clinicians to ask about current and past drug use and connect patients to services that have been proven to prevent premature death from drug-related causes.
To decrease mortality for people with HCV and decrease the proportion of people with HCV who die within three years of diagnosis, improvements in testing, care and treatment are needed. The younger age at HCV report in the co-infected group suggests that having an HIV diagnosis may prompt HCV testing, as is recommended [27] and, therefore, identification of the co-infected persons with HCV occurs slightly earlier, though not early enough to impact mortality. Despite the availability of up-to-date guidelines regarding HCV testing, a national study found that 42% of primary care physicians were unfamiliar with these guidelines; thus, they may be less likely to identify HCV infection their patients [28]. Implementing recommendations to test persons born between 1945 and 1965 [8], offering HCV antibody and RNA testing in primary care settings, using point-of-care antibody testing and RNA reflex
testing [10], and integrating HCV testing and treatment in high-risk settings could improve outcomes by identifying individuals with HCV at an earlier stage when steps such as limiting alcohol consumption to reduce risk of cirrhosis can be taken, and when treatment may be more effective [29]. This has been done successfully in methadone programs [30] and correctional settings [31].
With antiviral therapy, it is possible to cure most patients of HCV infection; cure is associated with decreases in all-cause and hepatitis-related mortality [32, 33]. Though barriers to accessing care remain and treatment completion rates are low [34, 35], recent improvements in HCV treatment and more expected in the coming years [11-13] will reduce barriers and more people may be cured. Even in the absence of treatment, earlier diagnosis is crucial, as physicians can promote liver health and prevent liver damage, for example through alcohol reduction counseling and vaccination against hepatitis A and B [29].
There are limitations to our analysis. Using the underlying cause of death recorded on the death certificate to ascribe cause-specific mortality may lead to potential misclassification of cause of death and bias towards some causes of death [36], e.g., HIV/AIDS. In addition, DOHMH vital statistics data only capture deaths occurring in NYC. Deaths that occurred out of jurisdiction were not captured, potentially underestimating the actual number of deaths. There are two additional limitations related to the use of HCV surveillance data. First, HCV surveillance may capture people who are not currently infected with HCV, because 15-25% of people that are antibody positive are RNA negative, either because of a resolved infection or a false-positive antibody result [1, 37]. Second, we identified some deaths due to HCV and HIV/AIDS that did not match to a HCV or HIV report in DOHMH surveillance registries. We did not add these individuals to the HIV or HCV categories, thus under-counting the infected population and the number of deaths from HIV/AIDS and HCV among persons with these diseases.
This analysis also has several strengths. It was the first population level analysis of cause specific mortality and HCV in NYC. The number of years of data allowed for a large sample size, which strengthens the findings. Many previous studies of HCV and mortality were limited to specific populations or cohorts [4, 16, 38]. The findings will inform DOHMH activities working with community providers to improve testing and treatment for HCV, and improve
outcomes for both HCV mono-infected and HCV/HIV co-infected persons. Further, matching surveillance registries with vital statistics mortality data may be replicable by other health departments to better understand their local patterns HCV mortality.
Identifying HCV infected persons earlier and linking them to comprehensive care and treatment services, and reducing new infections and preventing drug-related deaths through harm reduction is likely to have long and short-term benefits for reducing premature mortality and HCV-associated healthcare costs. The 2011 US Department of Health and Human Services Action Plan for the Prevention, Care and Treatment of Viral Hepatitis and the 2010 Institute of Medicine report both highlight the need to educate providers and communities; improve testing, care and treatment; improve surveillance and prevent new infections and mortality due to injection drug use [23, 29]. These efforts may reduce the number of infections and improve outcomes for those living with HCV. It is important to continue to track mortality among people with HCV to assess changes in age at death and causes of death, as these important developments in the public health response to HCV unfold.
|
|
| |
| |
|
|
|
