Cirrhosis and Liver Cancer Mortality in the United States · 2018. 7. 23. · Cirrhosis and Liver...
Transcript of Cirrhosis and Liver Cancer Mortality in the United States · 2018. 7. 23. · Cirrhosis and Liver...
Confidential: For Review Only
Cirrhosis and Liver Cancer Mortality in the United States
1999-2016: An Observational Study
Journal: BMJ
Manuscript ID BMJ.2018.043691.R1
Article Type: Research
BMJ Journal: BMJ
Date Submitted by the Author: 27-Apr-2018
Complete List of Authors: Tapper, Elliot; University of Michigan, Parikh, Neehar; University of Michigan
Keywords: liver disease, hepatitis, falls, alcohol, hepatocellular carcinoma
https://mc.manuscriptcentral.com/bmj
BMJ
Confidential: For Review Only
1
Cirrhosis and Liver Cancer Mortality in the United States 1999-2016: An
Observational Study
Elliot B. Tapper MD (1,2) and Neehar D Parikh MD MS (1,2)
1. Division of Gastroenterology and Hepatology, University of Michigan
2. Gastroenterology Section, VA Ann Arbor Healthcare System, Ann Arbor
Keywords: Liver disease, hepatitis, falls, alcohol, liver cancer
Word count: 3,439
Corresponding author:
Elliot B. Tapper, MD
3912 Taubman, SPC 5362
1500 E Medical Center Dr
Ann Arbor, MI 48109
T: (734) 647-9252
F: (734) 936-7392
Disclosure:
1. Elliot Tapper is the guarantor of this article
2. Roles
a. Concept: Tapper, Parikh
b. Analysis: Tapper, Parikh
c. Data acquisition: Tapper, Parikh
d. Writing: Tapper, Parikh
3. Conflicts of interest: no support from any organisation for the submitted work; no
financial relationships with any organisations that might have an interest in the submitted work
in the previous three years, no other relationships or activities that could appear to have
influenced the submitted work
4. Funding: Elliot Tapper receives funding from the National Institutes of Health through
the Michigan Institute for Clinical and Health Research (KL2TR002241).
5. The Corresponding Author has the right to grant on behalf of all authors and does grant on
behalf of all authors, a worldwide licence to the Publishers and its licensees in perpetuity, in all
forms, formats and media (whether known now or created in the future), to i) publish, reproduce,
distribute, display and store the Contribution, ii) translate the Contribution into other languages,
create adaptations, reprints, include within collections and create summaries, extracts and/or,
abstracts of the Contribution and convert or allow conversion into any format including without
limitation audio, iii) create any other derivative work(s) based in whole or part on the on the
Page 1 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
2
Contribution, iv) to exploit all subsidiary rights to exploit all subsidiary rights that currently exist
or as may exist in the future in the Contribution, v) the inclusion of electronic links from the
Contribution to third party material where-ever it may be located; and, vi) licence any third party
to do any or all of the above
6. The lead author affirms that the manuscript is an honest, accurate, and transparent account of
the study being reported; that no important aspects of the study have been omitted.
Page 2 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
3
Abstract
Objectives: Although both are often preventable, cirrhosis and hepatocellular carcinoma (HCC)
both characterized by the development of fatal, resource intensive complications. National data
regarding trends in demographics and associated causes of liver disease mortality are lacking.
We aimed to describe mortality by age-group, sex, race, and state of residence
Design: Observational cohort study
Setting: Death certificate data from the Vital Statistics Cooperative and population data from the
US Census Bureau compiled by the Center for Disease Control and Prevention’s Wide-ranging
Online Data for Epidemiologic Research (1999-2016).
Participants: US residents
Main outcome measures: Mortality. Trends evaluated using Joinpoint Regression.
Results: Deaths due to cirrhosis increased by 65% from 1999-2016 to 34,174; HCC deaths
doubled to 11,073. Only one demographic subgroup experienced improving HCC mortality, the
death-rate fell by 2.7% 95%CI(2.2-3.3,p<0.0001) per year for Asians/Pacific-Islanders. Annual
increases in cirrhosis mortality were most pronounces for American Indians (4.0% 9%CI(2.2-
5.7,p=0.002). The age-adjusted death-rate due to HCC both rose annually by 2.1% 95%CI(1.9-
2.3, p<0.0001); deaths due cirrhosis began rising in 2009 through 2016 by 3.4% 95%CI(3.1-
3.8,p<0.0001). Driven entirely by alcohol-related liver disease, persons aged 25-34 experienced
the highest average annual increase in cirrhosis mortality – 10.5% (95%CI8.9-12.2,p<0.0001) -
from 2009-2016. During this period, mortality due to peritonitis and sepsis in the setting of
cirrhosis rose substantially with respective annual increases of 6.1%(95%CI 3.9-8.2) and
7.1%(95%CI 6.1-8.4). Only one state, Maryland, had improving mortality (-1.2% (95%CI-1.7- -
0.7 per year), while many, concentrated in the South and West observed disproportionate annual
increases: Kentucky 6.8% (95%CI 5.1-8.5), New Mexico 6.0% (95%CI 4.1-7.9), Arkansas 5.7%
(95%CI 3.9-7.6), Indiana 5.0% (95%CI 3.8-6.1), and Alabama 5.0% (95% CI 3.2-6.8). No state
had improving HCC mortality while Arizona (5.1% 95%CI3.7-6.5) and Kansas (4.3%
95%CI2.8-5.8) saw the most severe annual increases.
Conclusions: Disparate changes in cirrhosis mortality by age, race, liver disease etiology, and
US region necessitate tailoring of preventative plans to inform rational public health outreach.
The rapid rise in death-rates among young persons due to alcohol highlight new challenges for
optimal care of patients with preventable liver disease.
Page 3 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
4
What this paper adds?
What is already known on this subject?
- Cirrhosis is a morbid condition that is increasing in prevalence
- Although hepatitis C is being eradicated, the burden of alcoholic liver disease and
Nonalcoholic Fatty Liver Disease is increasing
What this study adds
- Mortality due to cirrhosis is increasing in the US since 2009
- Persons aged 25-34 experienced the greatest relative increase in mortality driven by death
due to alcoholic cirrhosis.
- Specific demographic subgroups – namely Whites, American Indians and Hispanics –
have experienced the greatest increase in cirrhosis mortality.
- Cirrhosis mortality is improving in Maryland but worst in Kentucky, New Mexico, and
Arkansas
Page 4 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
5
Introduction
Cirrhosis, the end result for most chronic liver diseases,1 is characterized by fatal,
resource-intensive complications. These include hepatocellular carcinoma (HCC), hepatic
encephalopathy, gastrointestinal hemorrhage, infections, and renal failure. Compounding the
problem, the prevalence of cirrhosis in the United States (US) has doubled in the last decade.2 3
Although hepatitis C (HCV), a major cause of cirrhosis, could be eradicated given recently
widely available effective antiviral therapy, this trend is likely to continue,4 with deaths due to
cirrhosis are expected to triple by 2030.5 Projected increases in cirrhosis incidence are thought to
be driven primarily by the increasing prevalence of alcoholic liver disease and nonalcoholic fatty
liver disease (NAFLD).6-8 The impact of cirrhosis related morbidity and mortality is uneven
across the US with significant geographic and demographic variability.9 The trends in cirrhosis
related mortality in specific subgroups are important in understanding differential impact and
targets for improvement of cirrhosis related outcomes.
Cirrhosis and HCC carry personal, social, and financial burdens for patients, their
families, and society.10-12
Prevention and early management of cirrhosis is therefore cost-
effective.13-15
However, nationally representative data to guide the optimal allocation of
resources and preventive efforts are limited, particularly regarding differential mortality by
demographic subgroups or regions of the country.16 Herein, to fill this gap, we analyze the
Center for Disease Control and Prevention’s Wide-ranging Online Data for Epidemiologic
Research (CDC WONDER) platform.
Page 5 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
6
Methods
All data were obtained from the CDC WONDER platform.17 For more details on CDC
Wonder please see the Supplemental Material. Our primary aim was to describe temporal
trends in death-rates attributable to cirrhosis (ICD-10 K70.3, K74.5, and K74.6) and HCC
(C22.0) as the primary/underlying cause of death for US adults. All rates obtained were age-
adjusted, in which age-specific mortality is weighted according to the age distribution in a
standard year (2000).18 Secondarily, we sought to describe how these trends differed based on
demographic subgroups; age, sex, race (Asian Pacific Islander, American Indian/Alaska Native,
black/African American, and white/Caucasian), Hispanic ethnicity, and place of residence. We
also aimed to describe trends in causes of death related to specific complications associated with
cirrhosis. Specifically, we examined death due to gastrointestinal hemorrhage (ICD-10 K25-K28,
K92.0-K92.2, I85.0), peritonitis (K65), sepsis (A41), hepatorenal syndrome (K76.7), or traumas
(V01-Y89) in patients with cirrhosis. Given the association between NAFLD and complications
of the metabolic syndrome like cerebrovascular disease and ischemic heart disease (I20-I25, I60-
I69).19 We also compared trends in deaths due to alcohol use disorder (F10).
To test for unmeasured secular trends, all findings were compared to death-rates due to
causes other than cirrhosis. Specifically, we examined the trends in deaths due to infections
(ICD-10 A00-B99), neoplasia (C00-C48, not including HCC, C22), cardiovascular disease (I00-
I99), and respiratory disease (J00-J98). Finally, we assessed overall trends in several states that
were observed to have the greatest increase or highest death-rates due to cirrhosis and HCC (e.g.
Arizona, New Mexico, and Wyoming).
Page 6 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
7
We performed several sensitivity analyses. First, we evaluated trends when the cause of
death was labelled as any liver disease with or without cirrhosis (e.g. viral hepatitis, ICD-10
K70-K76). Second, we evaluated trends when the death certificate included any mention
cirrhosis as a primary or contributory cause. A death certificate can have a primary/underlying
cause and up to 20 contributing causes. Third, we repeated trend analysis that included age-
adjusted death-rate estimates by standardizing rates to different years (2000, 2009, and 2010) and
gender (men/women).
Data Analysis
Trends in death-rates were then evaluated using the National Cancer Institute’s Joinpoint
program (version 4.6.0; available: http://surveillance.cancer.gov/joinpoint). This technique
allows us to identify if there were years in the study period where the rate of change in the death-
rate was significantly different. It, which uses a piecewise linear regression approach to
determine whether rates over time are best described by a straight line (0 joinpoints) or by
multiple linear segments (>1 joinpoints).20 21
The software detects the joinpoints in the trends
using grid search method of observe the joinpoints at any point in the time trend,22 fitting a linear
regression between consecutive joinpoints using the least squares approach. To simplify the
model, we specified modeling procedures to keep a maximum of 3 joinpoints (changes in slope)
with a minimum of 3 years per segment. Using validated methodology, we obtained the annual
percentage change (APC) in death-rates over each linear segment as well as the average annual
percentage change (AAPC) with 95% confidence intervals (CI) from 1999-2016 for each
model.23 24
The APC was obtained using a Monte-Carlo permutation analysis. Given the
uncertainty (within the confidence intervals) of each year’s mortality rate, this analytic approach
Page 7 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
8
fits a line through 4,500 different samples of each year’s rate in order to estimate the APC. The
permutation test provides a p-value reflecting the probability that the original sample based on
the calculated mortality estimates were due to chance. The software selects a final model based
on the minimum number of joinpoints (within the set limits) to explain the mortality trends. All
p-values were adjusted through a Bonferroni correction. The AAPC is a weighted average of the
segmented annual percent change (APC) over a specified time interval, providing a valid
summary statistic to compare rate changes across groups. We found one Joinpoint in the trends
of cirrhosis mortality at 2009. To further understand demographic trends, we analyzed our data
from 2009-2016.
When comparing the trends in mortality due to cirrhosis with control conditions, we
evaluated the differences using a test for parallelism.25 This test evaluates whether curves of
differing absolute values are parallel over multiple (in this case 4,500) permutations of each
model. Patients were not involved in the study design. As this is a publically available dataset of
de-identified data, no ethics approval was obtained.
Page 8 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
9
Results
Burden of Mortality Due to Cirrhosis and HCC
During the study period, cirrhosis caused a total of 460,760 deaths (20,661 in 1999 and
34,174 in 2016) and 136,442 were attributed to HCC (5,112 in 1999 and 11,073 in 2016).(Table
1) Males had a higher burden of age-adjusted cirrhosis mortality when compared to females by a
2:1 ratio and a higher burden of HCC related deaths by a nearly 4:1 ratio. The highest burden of
age-adjusted cirrhosis mortality was seen in American Indians (25.8/100,000 people), followed
by whites (12.7/100,000 people). Age-adjusted HCC death-rates were highest in Asian/Pacific
Islander patients. Compared to non-Hispanic persons, Hispanics had relatively high age-adjusted
mortality from both cirrhosis (19.3 vs 11.5/100,000) and HCC (5.7 vs 3.4/100,000). The
Southern and Western regions of the US had the highest age-adjusted mortality from cirrhosis
(13.5/100,000 and 13.3/100,000, respectively) while the Western US alone had the highest age
adjusted mortality related to HCC (4.2/100,000). In Supplementary Tables 1a-1b we provide
death rates due to all liver disease codes (1a) and deaths where cirrhosis was noted in any of the
up-to-twenty positions on the death certificate (1b). While these sensitivity tests accounted for
more deaths (742,800 and 800,344 respectively), the trends were the same with statistically
parallel annual trends for each subgroup. In Supplementary Table 1c, we provide death-rates
for persons with both cirrhosis and HCC on the same death certificate. Although the number of
deaths are lower, the relative burden by subgroup is similar to HCC overall with the exception
that American Indians have a relatively higher burden of HCC and cirrhosis.
Page 9 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
10
Demographic Trends in Mortality Due to Cirrhosis and HCC
Cirrhosis
We evaluated trends in cirrhosis death-rates in demographic subgroups with several
significant findings.(Table 2) First, there was an inflection point at 2009 where cirrhosis-related
mortality increased significantly through to 2016. Several subgroups had significant declines in
cirrhosis-related mortality from 1999-2008, however these trends reversed in nearly every
demographic subgroup beginning in 2009. Second, although more men are dying from (or with)
cirrhosis and HCC, both sexes experienced similar rates of increased mortality over time. Third,
whites and American Indians had rapid increases in cirrhosis mortality after 2009 with AAPC of
3.3%(95%CI 2.8-3.7) and 4.0%(95%CI 2.2-5.7) compared to African Americans. Fourth, non-
Hispanic patients experienced a slower rise in cirrhosis related mortality than Hispanic patients.
Fifth, in recent years, the Northeast US had the slowest rise (AAPC 1.6% 95%CI 1.1-2.1) while
the south had the fastest (3.5% 95%CI 2.8-4.2).
Hepatocellular Carcinoma
The demographic trends in mortality from HCC are markedly different from cirrhosis.
First, overall, there was a 2.1% (95%CI:1.9-2.3) increase in HCC mortality nationally over the
study period. Second, females experienced a consistent increase in HCC mortality, while HCC
death-rates for males slowed beginning in 2011. Third, Asian/Pacific Islanders were the only
subgroup with an improvement in HCC mortality, experiencing a consistent decrease of 2.7%
(95%CI 2.2-3.3) per year. Third, all US regions except for the Northeast experienced significant
increases HCC mortality over the study period.
Page 10 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
11
Age-Related Trends in Cirrhosis and HCC Mortality
We performed several analyses in order to better delineate the populations affected by
worsening trends in cirrhosis mortality during the period of worsening death-rates (2009-2016).
First, we examined changing death-rates by age-group.(Figure 1) Persons aged 25-34
experienced the highest AAPC in cirrhosis mortality; 3.7% (95%CI2.4-5.1) over the whole
period and 10.5% (95%CI8.9-12.2) from 2009-2016.(Figure 1) Whites and American Indians
had the most rapid increase in this age group. Increased death-rates in the 25-34 age group were
driven by alcoholic cirrhosis, where the average AAPC from 2009-2016 was 10.5% (95%CI 9.2-
11.8). These findings are reinforced by parallel changes mortality due to alcohol use disorder and
all alcohol-related liver diseases (Panel D). In Supplementary Tables 2a-2b, we show the
proportion of all-cause mortality has varied by age group and race 2009-2016. Notably, we show
that by 2016, cirrhosis accounts for 6.3% (up from 4.3% in 2009) and 7.0% (from 5.8% in 2009)
of deaths for American Indians aged 25-34 and >35 respectively. Cirrhosis accounted for 1.4%
(up from 0.9%) and 2.3% (up from 1.8%) of deaths for persons age 25-34 and >35 respectively.
Finally, as seen in Figure 1A, there is an opposite age relationship with respect to HCC-related
mortality; it has decreased in younger patients (<55 years) since 2009, and increased among
those >55 years-old from 2009-2016. In a sensitivity test, we evaluated patients with both HCC
and cirrhosis on the same death certificate. A formal trend analysis of young persons was not
possible as 80% of cirrhosis-HCC deaths occurred in persons older than 55. Only 58 deaths were
recorded for this combination of causes for persons aged 25-34.
Cause-Specific Mortality Trends
Page 11 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
12
Second, we sought to determine whether specific common causes of death in patients
with cirrhosis were changing over the study period. Supplementary Table 3 depicts the time
trends in comorbid causes of death for patients who died from cirrhosis. The AAPC for
cirrhosis-related deaths due to peritonitis (1.5% 95%CI0.2-2.8) and sepsis (3.7% 95%CI3.2-4.2)
increased significantly while no significant changes were observed for comorbid upper
gastrointestinal hemorrhage, hepatorenal syndrome, cerebrovascular and ischemic heart disease,
and trauma. However, since 2009, the mortality associated with each condition increased save
for trauma. Notably, from 2009-2016 the magnitude of increased death-rate due to peritonitis and
sepsis rose substantially with respective AAPC of 6.1%(95%CI 3.9-8.2) and 7.1%(95%CI 6.1-
8.4). Most (60%) of the 47,290 patients with sepsis or peritonitis were aged 45-64. All groups
had increasing mortality, but the rate of change was not statistically parallel. For example,
persons aged 25-34 and 35-44 experienced sharp increases in sepsis/peritonitis related mortality
beginning in 2009; the AAPC (2009-2016) for these groups were 13.5% 95%CI(9.4–17.4) and
7.4% 95%CI(4.3–10.6). Persons aged 75-84 experienced a later joinpoint with an increase in
sepsis/peritonitis mortality from 2013-2016, APC 12.4% 95%CI(3.8-21.8) and persons aged >85
years had a linear increase from 1999-2016 of 3.3% 95%CI(2.0-4.6).
State-Level Variation in Mortality
We next examined geographic changes in cirrhosis and HCC-related mortality since
2009. The fastest rise in cirrhosis and HCC related mortality occurred in the Western and
Southern US. In Supplementary Tables 4-5, we detail the state-level changes in mortality from
2009-2016; In Figure 2 we show in a series of heat-maps how the death-rate due to cirrhosis and
HCC has changed over time.
Page 12 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
13
In Figure 3A, we focus on the states with the largest increases in mortality due to
cirrhosis. The states with the greatest AAPC in cirrhosis related mortality include Kentucky
6.8% (95%CI 5.1-8.5), New Mexico 6.0% (95%CI 4.1-7.9), Arkansas 5.7% (95%CI 3.9-7.6),
Indiana 5.0% (95%CI 3.8-6.1), and Alabama 5.0% (95% CI 3.2-6.8). One state, Maryland,
experienced a significant annual decrease in cirrhosis related deaths, -1.2% (95%CI-1.7- -0.7).
No state experienced a significantly decreasing mortality due to HCC (Figure 3B). Many
of the same states with worsening cirrhosis mortality also experienced worsening HCC mortality
(e.g. Oregon, Iowa). However, some states experienced increased HCC-related deaths out-of-
sync with cirrhosis trends. These include Arizona (5.1% 95%CI3.7-6.5), Kansas (4.3%
95%CI2.8-5.8), Kentucky (4.0% 95%CI3.1-5.0), and Washington (3.9% 95%CI3.1-4.6).
State-Level Variation in Mortality Due to Alcohol-Related Diseases
In Figure 4, we evaluated the state-level trends in alcohol-related mortality. In 4A, we
show that the AAPC for deaths due to alcoholic cirrhosis increased most rapidly in Louisiana
(13.1% 95%CI 7.7-18.9), California (12.9% 95%CI6.7-19.5), Kentucky (12.8% 95%CI9.8-15.8),
Delaware (9.8% 95%CI5.1-14.7), and New Mexico (9.3% 95%CI6.3-12.4). Conversely, the
District of Columbia and Maryland had the only improvements in mortality. In 4B, we show
trends in the AAPC for deaths due to alcohol use disorder. None of the state-level trends in
deaths due to alcohol use disorder are statistically parallel with those of alcohol-related cirrhosis,
however many are qualitatively similar. In Supplementary Table 6, we show the age-adjusted
mortality and annual changes in mortality for alcohol use disorder for each state.
Page 13 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
14
Secular Trends in Non-cirrhotic death-rates (1999-2016)
We examined overall death-rates in each state. Age-adjusted mortality is uniformly
decreasing across the states with the exception of Utah, where the AAPC 2009-2016 was 0.6%
95%CI(0.3 – 0.8). A sample of states (chosen for their significant changes in cirrhosis mortality)
are shown in Supplementary Table 7. None of the state-wide all-cause mortality trends were
statistically parallel with those of cirrhosis or HCC. For example, where cirrhosis-related
mortality improved in Maryland beginning in 2009, all-cause mortality improved in Maryland
until 2010 when it plateaued.
We then evaluated trends in deaths due to other causes over the study period. In
Supplementary Table 8, we compared trends in cirrhosis-related mortality to deaths due to
infections, neoplasia, cardiovascular disease, and respiratory diseases to assess for the presence
of unmeasured secular trends. First, all other disease categories were associated with
improvements in mortality while cirrhosis worsened. Second, although each category improved,
the trends observed for the other causes were not statistically parallel. Third, each category was
associated with continuous (linear) improvement except infections, which began to improve most
after 2009.
Page 14 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
15
Discussion
Cirrhosis and HCC are deadly conditions, each with increasing footprints in American
public health. As death is one outcome from advanced liver disease among others including
caregiver burden and disability, increasing mortality due to cirrhosis and HCC speak to the
broadening socioeconomic impact of liver disease. Adverse trends in liver-related mortality are
particularly unfortunate given that in most cases the liver disease is preventable. Understanding
the factors associated with mortality due to these conditions will inform how best to allocate
resources. In this study of national data from 1999-2016, we find several major trends in
cirrhosis and HCC mortality in the US.
Worsening Cirrhosis and HCC Mortality
Mortality due to cirrhosis has been rising since 2009. Asrani et al previously
demonstrated unchanged age-and-sex-adjusted death-rates due to liver disease from 1979
(25.8/100,000) to 2008 (25.7/100,000) in Olmstead county.26 Our analysis demonstrates
increased mortality from 1999-2016 in the context of declining mortality due to other causes
such as cardiovascular disease, neoplasia, respiratory disease, and infectious diseases.
Specific Demographics Are Disproportionately Affected
Increases in cirrhosis mortality disproportionately affects young persons aged 25-34,
White, American Indian, and Hispanic Americans. These changes appear to be driven by
alcohol-related liver disease and alcohol-use disorder, which are largely preventable causes of
Page 15 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
16
death in this age group. Previous estimates of alcohol-related diseases (prior to 2010) show that
mortality was concentrated in persons aged >65 years.27 This study extends such data by
showing the rate-of-change is greatest for the young. The impact of these preventable deaths is
compounded by their economic impact, measured in decades per person of lost productivity and
outsized healthcare costs (twice that of nonalcoholic cirrhosis8). Conversely, mortality due to
HCC continues to increase in older Americans, it is decreasing in younger people and Asians.
These trends also parallel those observed for the incidence of HCC.9 Potential explanations
supported by these data include increasing early detection of HCC, application of
curative/locoregional therapies, and, because hepatitis B is the principal cause of HCC
worldwide and among Asian-Americans, effectiveness of vaccination programs and efficacy in
antiviral therapy for hepatitis B in preventing development of HCC.28-31
Infections Are a Target for Prevention
The causes of death associated with cirrhosis that had the most consistent and substantial
increases over the study period were peritonitis and sepsis. Patients with cirrhosis are at
increased risk of infections due to impaired immunity and propensity for bacterial translocation,
especially in the setting of portal hypertension.10 32-34
Infections, particularly spontaneous
bacterial peritonitis (SBP), are a major cause of death in cirrhosis patients and are major
contributors to our observed rise in cirrhosis related mortality. Proton pump inhibitors, both
over-prescribed and associated with SBP as well as increased susceptibility to antibiotic resistant
bacterial organisms, may contribute to the observed trends.35
Page 16 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
17
State-Level Variations in Mortality
Specific states in the South and West experienced the worst trends for cirrhosis and HCC
mortality. Only Maryland claimed consistent reductions in cirrhosis mortality. Adverse trends in
Kentucky, New Mexico, Arizona, and Oregon in particular require further study for explanation.
In seeking explanations, given that worsening trends began after 2008, the global financial crisis
and the subsequent economic recession in the US is a possible candidate which may in part
explain the outsized impact in some states. This dovetails with trends affecting younger men:
becoming unemployed is linked with alcohol abuse in young men but not older persons or
women.36
Clinical and Health-Policy Implications
Cirrhosis is morbid and costly to treat and therefore efforts to prevent progression of all
liver diseases are cost-effective.14 15 37 38
Our data highlight several modifiable sources of
mortality. First, young persons, particularly American Indians and Whites, may benefit from
intensified resources for screening for and treatment of alcohol use disorders. Second, quality
improvement efforts are needed optimize the identification and early-goal directed therapy of
SBP and decrease the overuse of proton pump inhibitors;39 additional data are needed to identify
the optimal preventative strategies for SBP in the era of multidrug resistant organisms. Third,
research and outreach are required to understand and contextualize the factors associated with
the adverse trends identified in many states. In addition to resources for alcohol-related disease,
the role of state-level policy instruments such as increased alcohol taxes (which have been linked
with decreased alcohol-related disease40 41
) deserve attention.
Page 17 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
18
Contextual Factors
Several considerations must be made in the interpretation of these findings. First, these
findings are based on data abstracted from death certificates. Death certificates may be
inaccurate (in up to 15% of cases42) and this study does not explicitly validate disease coding
with chart review. For this reason, we chose the diagnostic codes most specific to cirrhotic
complication. Still, sensitivity analyses using a variety of coding algorithms for liver disease
confirm consistency of results and comparisons of our results to other conditions imply
independence from secular trends. Second, though we have detected worsening mortality since
2009, the precise explanations for this trend and the geographic heterogeneity in our analysis
require further study. For example, we identify the most at-risk states, but granular data is needed
to determine root-causes, be they trends in healthcare access and insurance, policy (e.g. alcohol
tax-rates or laws), employment, education levels, or the differential changes in the comorbidities
(e.g. obesity) of a given state’s population.
Third, it is unclear how these trends are – or will be – affected direct-active antivirals for
HCV. We have included 3 years of data since the approval of DAAs and the observed trends
have remained consistent. Screening for HCV is recommended given the availability of highly
efficacious therapy, however hundreds of thousands of individuals in the US have undetected
HCV or do not have access to antiviral therapy.15 43
At the same time, HCV eradication will
prevent the development of cirrhosis and its complications, potentially changing these trends if
re-assessed in the next 5-10 years. However, HCV therapy cannot modify the significant trends
observed related to alcohol or the expected increase in the burden of NAFLD.5
Page 18 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
19
Conclusion
Specific demographic subgroups and states are disproportionately affected by adverse
trends in cirrhosis and HCC-related mortality. Additionally, patients with cirrhosis are more
likely to die due to specific secondary causes (i.e. infections.) These data underscore gaps in care
and opportunities for prevention.
Page 19 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
20
Table 1: Overall Demographics 1999-2016
Total
s and rates are cumulative and reflect the entire period 1999-2015.PI = pacific islander.
Cirrhosis Deaths Hepatocellular Carcinoma Deaths
Deaths Crude rate
(per 100,000) Age-adjusted (per 100,000) Deaths
Crude rate (per 100,000)
Age-adjusted (per 100,000)
Overall 460,760 12.87 12.83 12.91 12.18 12.14 12.21 136,442 3.81 3.79 3.83 3.55 3.53 3.56
Female Male
162,599 298,161
8.75 8.71 8.79 8.02 7.98 8.06 31,694 105,248
1.70 1.68 1.71 1.50 1.48 1.51
17.31 17.25 17.37 16.82 16.76 16.88 6.1 6.06 6.14 5.97 5.93 6.01
American Indian Asian/PI
African American White
9,145 6,687 40,598
404,330
23.7 23.21 24.18 25.81 25.26 26.37 1,386 11,241 21,303
103,012
3.6 3.41 3.79 4.72 4.46 4.99
3.59 3.51 3.68 4.45 4.34 4.56 6.02 5.91 6.13 7.43 7.29 7.57
9.43 9.33 9.52 9.71 9.61 9.81 4.94 4.87 5.00 5.20 5.13 5.27
13.82 13.78 13.86 12.71 12.67 12.75 3.51 3.49 3.53 3.15 3.13 3.17
Hispanic Non-Hispanic
64,001 395,128
14.18 14.07 14.29 19.27 19.11 19.42 17,288 119,250
3.82 3.76 3.88 5.74 5.65 5.83
12.63 12.59 12.67 11.51 11.47 11.54 3.80 3.78 3.82 3.37 3.35 3.39
Northeast Midwest South West
71,388 90,958
187,949 110,465
10.65 10.58 10.73 9.81 9.74 9.89 23,996 24,172 45,105 32,596
3.85 3.8 3.90 3.48 3.43 3.52
11.60 11.52 11.67 10.78 10.71 10.85 3.34 3.30 3.38 3.06 3.02 3.10
14.32 14.25 14.38 13.53 13.47 13.6 3.75 3.72 3.79 3.49 3.46 3.52
13.58 13.5 13.66 13.33 13.25 13.41 4.34 4.29 4.38 4.23 4.19 4.28
Page 20 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
21
Table 2: Trends in the Annual Percentage Change in Death-rates Due to Cirrhosis and Hepatocellular Carcinoma
Time Trends in Cirrhosis Deaths Time Trends in Hepatocellular Carcinoma Deaths
Years of
similar trend APC (95%CI) P-value Years of
similar trend APC (95%CI) P-value
Overall 1999-2008 2008-2016
-0.5 (-0.9- -0.1) 3.4 (3.1-3.8)
0.02 <0.00001 Overall 1999-2016 2.1 (1.9-2.3) <0.00001
Female Female
Male Male
1999-2008 2008-2016 1999-2009 2009-2016
-0.3 (-0.8-0.2) 3.6 (3.1-4.2)
-0.7(-1.1- -0.3) 2.6 (2.1-3.0)
0.2 <0.00001
0.003 <0.00001
Female Male Male
1999-2016 1999-2010 2010-2016
1.3 (1.9-1.7) 2.6 (2.1-3.1) 1.3 (0.4-2.3)
<0.00001 <0.0001
0.02
American Indian American Indian
Asian/PI African American African American
White White
1999-2010 2010-2016 1999-2016 1999-2008 2008-2016 1999-2008 2009-2016
-0.3(-1.2-0.5) 4.0 (2.2-5.7) 0.3 (-0.2-0.7)
-3.6 (-4.3- -3.9) 1.7 (0.9-2.4)
-0.1 (-0.5-0.3) 3.3 (2.8-3.7)
0.4 0.002
0.2 <0.00001
0.003 0.7
<0.00001
American Indian Asian/PI
African American White
1999-2016 1999-2016 1999-2016 1999-2016
2.3 (1.1-3.5) -2.7 (-3.3- -2.2)
2.2 (1.7-2.7) 2.4 (2.2-2.8)
0.007 <0.00001 <0.00001 <0.00001
Hispanic Hispanic
Non-Hispanic Non-Hispanic
1999-2007 2008-2016 1999-2008 2009-2016
-1.7 (-2.3 - -1.0) 1.3 (0.8 – 1.7)
-0.5 (-0.9 - -0.1) 3.0 (2.5 – 3.4)
0.0003 <0.00001
0.02 <0.00001
Hispanic Non-Hispanic Non-Hispanic Non-Hispanic Non-Hispanic
1999-2016 1999-2006 2006-2010 2010-2014 2014-2016
0.7 (0.02-1.2) 2.0 (1.3-2.7) 3.3 (1.1-5.6)
0.5 (-1.5 – 2.6) 3.7 (-0.2 – 7.7)
0.009 0.0004 0.01 0.1
Northeast Northeast Midwest Midwest South South West West
1999-2007 2007-2016 1999-2008 2008-2016 1999-2009 2009-2016 1999-2008 2008-2016
-2.2 (-2.8- -1.6) 1.6 (1.1 - 2.1)
-0.4 (-0.7- -0.1) 3.1 (2.7-3.4) -0.1(-0.6-0.3) 3.5 (2.8-4.2) -0.1(-0.7-0.6) 3.0 (2.3-3.6)
0.0003 0.0008 0.01
<0.00001 0.4
<0.00001 0.8
<0.00001
Northeast Northeast Midwest South West
1999-2010 2010-2015 1999-2015 1999-2015 1999-2015
2.5 (1.6-3.4) 0.7 (-0.2-1.6) 2.5 (2.3-2.7) 2.3 (1.9-2.8) 1.8 (1.5-2.1)
0.002 0.1
<0.00001 <0.00001 <0.00001
In this table, an analysis of the trends in the changing rates of death are shown. Each segment shown is derived from a Joinpoint analysis where
the years included in the trend are chosen based on similarities in the annual rate of change. For example, if the whole study period (1999-2016)
is included, it is because there are no periods > 3 years in which the rate of change is significantly different. AI = American Indian, A/PI =
Asian/Pacific Islander, AA = African American, NE = Northeast, MW = Midwest, S = South, W = West
Page 21 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
22
Figure 1: Death-rate Trends by Age-Group 2009-2016
This graph explores changes in mortality by subgroup over the period in which there was an
overall increase in cirrhosis mortality. Each bar represents the average annual percentage change
in the death-rate due to cirrhosis and hepatocellular carcinoma (HCC) in all adult age groups
from 2009-2016. Panel A depicts changes in overall cirrhosis and HCC mortality; B shows
changes by race; C shows changes by sex; and D shows changes by alcohol-related diagnosis
codes.
Figure 2: Age Adjusted Death-rates Attributable to Cirrhosis and Hepatocellular
Carcinoma over Time
The age-adjusted mortality for each condition in each state is presented in quartiles for the first,
middle and last year of the study. Quartiles for each time period are denoted by yellow (first),
light orange (second), dark orange (third), and red (fourth). Striped or shaded states imply data
that is either unreliable or suppressed to protect patient identity.
Figure 3: States with Significant changes in Death-rates Attributable to Cirrhosis and
Hepatocellular Carcinoma (HCC) 2009-2016
This graph explores the changes in cirrhosis and HCC mortality at the state level 2009-2016. The
average annual percentage change (AAPC) in mortality for each individual state is plotted
against the age-adjusted death-rate in 2009. States without significant trends were excluded from
this figure for simplicity. Panel A shows trends for cirrhosis and B shows trends for HCC.
Figure 4: Average Annual Percentage Change (AAPC) in Mortality Due to Alcohol-Related
Disease
Note: Many states were excluded from the figures for simplicity because their trends were non-
significant (no changes over the specified timeframes).
Top: The AAPC for mortality by state for alcohol-related cirrhosis (ICD-10 K70.3) is shown for
2009-2016.
Bottom: The AAPC for mortality by state for alcohol-use disorder (ICD-10 F10) is shown for
2009-2016.
Page 22 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
23
References
1. Schuppan D, Afdhal NH. Liver cirrhosis. Lancet 2008;371(9615):838-51.
2. Beste LA, Leipertz SL, Green PK, et al. Trends in burden of cirrhosis and hepatocellular carcinoma by
underlying liver disease in US veterans, 2001-2013. Gastroenterology 2015;149(6):1471-82 e5;
quiz e17-8. doi: 10.1053/j.gastro.2015.07.056
3. D'Amico G, Garcia-Tsao G, Pagliaro L. Natural history and prognostic indicators of survival in cirrhosis:
a systematic review of 118 studies. J Hepatol 2006;44:217-31. doi: 10.1016/j.jhep.2005.10.013
[published Online First: 2005/11/22]
4. Moon A, Green P, Berry K, et al. Transformation of hepatitis C antiviral treatment in a national
healthcare system following the introduction of direct antiviral agents. Alimentary
Pharmacology & Therapeutics 2017;45(9):1201-12.
5. Estes C, Razavi H, Loomba R, et al. Modeling the epidemic of nonalcoholic fatty liver disease
demonstrates an exponential increase in burden of disease. Hepatology
6. Parikh ND, Marrero WJ, Wang J, et al. Projected increase in obesity and non-alcoholic steatohepatitis-
related liver transplantation waitlist additions in the United States. Hepatology 2017
7. Younossi ZM, Stepanova M, Afendy M, et al. Changes in the prevalence of the most common causes
of chronic liver diseases in the United States from 1988 to 2008. Clinical Gastroenterology and
Hepatology 2011;9(6):524-30. e1.
8. Mellinger JL, Shedden K, Winder GS, et al. The High Burden of Alcoholic Cirrhosis in Privately Insured
Persons in the United States. Hepatology 2018
9. White DL, Thrift AP, Kanwal F, et al. Incidence of hepatocellular carcinoma in all 50 United States,
from 2000 through 2012. Gastroenterology 2017;152(4):812-20. e5.
10. Tapper EB, Finkelstein D, Mittleman MA, et al. Standard assessments of frailty are validated
predictors of mortality in hospitalized patients with cirrhosis. Hepatology 2015;62(2):584-90.
11. Bajaj JS, Wade JB, Gibson DP, et al. The multi-dimensional burden of cirrhosis and hepatic
encephalopathy on patients and caregivers. The American journal of gastroenterology
2011;106(9):1646-53.
12. Patel AA, Walling AM, May FP, et al. Palliative Care and Health Care Utilization for Patients With End-
Stage Liver Disease at the End of Life. Clinical Gastroenterology and Hepatology 2017
13. Tapper EB, Catana AM, Sethi N, et al. Direct costs of care for hepatocellular carcinoma in patients
with hepatitis C cirrhosis. Cancer 2016;122(6):852-58.
14. Tapper EB, Sengupta N, Hunink MM, et al. Cost-effective evaluation of nonalcoholic fatty liver
disease with NAFLD fibrosis score and vibration controlled transient elastography. The American
journal of gastroenterology 2015;110(9):1298.
15. Rein DB, Smith BD, Wittenborn JS, et al. The cost-effectiveness of birth-cohort screening for hepatitis
C antibody in US primary care settings. Annals of internal medicine 2012;156(4):263-70.
16. White DL, Thrift AP, Kanwal F, et al. Incidence of Hepatocellular Carcinoma in All 50 United States,
From 2000 Through 2012. Gastroenterology 2017;152(4):812-20 e5. doi:
10.1053/j.gastro.2016.11.020
17. Aghaei I, Saeedi Saravi SS, Ghotbi Ravandi S, et al. Evaluation of prepulse inhibition and memory
impairments at early stage of cirrhosis may be considered as a diagnostic index for minimal
hepatic encephalopathy. Physiology & behavior 2017;173:87-94. doi:
10.1016/j.physbeh.2017.01.033 [published Online First: 2017/01/26]
18. Anderson RN, Rosenberg HM. Age standardization of death rates: implementation of the year 2000
standard. National vital statistics reports;47(3):1-17.
19. Allen AM, Terry TM, Larson JJ, et al. Nonalcoholic Fatty Liver Disease Incidence and Impact on
Metabolic Burden and Death: a 20 Year-Community Study. Hepatology 2017
Page 23 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
24
20. Kim HJ, Fay MP, Feuer EJ, et al. Permutation tests for joinpoint regression with applications to cancer
rates. Statistics in medicine 2000;19(3):335-51.
21. Hudson DJ. Fitting segmented curves whose join points have to be estimated. Journal of the
american statistical association 1966;61(316):1097-129.
22. Lerman P. Fitting segmented regression models by grid search. Applied Statistics 1980:77-84.
23. Kim HJ, Luo J, Chen HS, et al. Improved confidence interval for average annual percent change in
trend analysis. Statistics in medicine 2017;36(19):3059-74.
24. Clegg LX, Hankey BF, Tiwari R, et al. Estimating average annual per cent change in trend analysis.
Statistics in medicine 2009;28(29):3670-82.
25. Kim HJ, Fay MP, Yu B, et al. Comparability of segmented line regression models. Biometrics
2004;60(4):1005-14.
26. Asrani SK, Larson JJ, Yawn B, et al. Underestimation of liver-related mortality in the United States.
Gastroenterology 2013;145(2):375-82. e2.
27. Rehm J, Samokhvalov AV, Shield KD. Global burden of alcoholic liver diseases. Journal of hepatology
2013;59(1):160-68.
28. Chen C-J, Yang H-I, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum
hepatitis B virus DNA level. Jama 2006;295(1):65-73.
29. Altekruse SF, McGlynn KA, Reichman ME. Hepatocellular carcinoma incidence, mortality, and survival
trends in the United States from 1975 to 2005. J Clin Oncol 2009;27(9):1485-91.
30. Altekruse SF, Henley SJ, Cucinelli JE, et al. Changing hepatocellular carcinoma incidence and liver
cancer mortality rates in the United States. Am J Gastroenterol 2014;109(4):542-53. doi:
10.1038/ajg.2014.11
31. Ulahannan SV, Duffy AG, McNeel TS, et al. Earlier presentation and application of curative
treatments in hepatocellular carcinoma. Hepatology 2014;60(5):1637-44. doi:
10.1002/hep.27288
32. Tapper EB, Risech-Neyman Y, Sengupta N. Psychoactive medications increase the risk of falls and fall-
related injuries in hospitalized patients with cirrhosis. Clinical Gastroenterology and Hepatology
2015;13(9):1670-75.
33. Lai JC, Feng S, Terrault NA, et al. Frailty predicts waitlist mortality in liver transplant candidates.
American journal of transplantation 2014;14(8):1870-79.
34. Bajaj JS, O’Leary JG, Tandon P, et al. Hepatic Encephalopathy Is Associated With Mortality in Patients
With Cirrhosis Independent of Other Extrahepatic Organ Failures. Clinical Gastroenterology and
Hepatology 2017;15(4):565-74. e4.
35. Liver EAftSot. EASL Clinical Practice Guidelines for the management of patients with decompensated
cirrhosis. Journal of hepatology 2018
36. Temple MT, Fillmore KM, Hartka E, et al. A meta-analysis of change in marital and employment
status as predictors of alcohol consumption on a typical occasion. Addiction 1991;86(10):1269-
81.
37. Leidner AJ, Chesson HW, Spradling PR, et al. Assessing the effect of potential reductions in non-
hepatic mortality on the estimated cost-effectiveness of hepatitis C treatment in early stages of
liver disease. Applied health economics and health policy 2017;15(1):65-74.
38. Veldhuijzen IK, Toy M, Hahné SJ, et al. Screening and early treatment of migrants for chronic
hepatitis B virus infection is cost-effective. Gastroenterology 2010;138(2):522-30.
39. Tapper EB. Building effective quality improvement programs for liver disease: a systematic review of
quality improvement initiatives. Clinical Gastroenterology and Hepatology 2016;14(9):1256-65.
e3.
40. Ponicki WR, Gruenewald PJ. The impact of alcohol taxation on liver cirrhosis mortality. Journal of
studies on alcohol 2006;67(6):934-38.
Page 24 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
25
41. Xuan Z, Chaloupka FJ, Blanchette JG, et al. The relationship between alcohol taxes and binge
drinking: evaluating new tax measures incorporating multiple tax and beverage types. Addiction
2015;110(3):441-50.
42. German RR, Fink AK, Heron M, et al. The accuracy of cancer mortality statistics based on death
certificates in the United States. Cancer epidemiology 2011;35(2):126-31.
43. Chhatwal J, Wang X, Ayer T, et al. Hepatitis C Disease Burden in the United States in the era of oral
direct-acting antivirals. Hepatology 2016;64(5):1442-50. doi: 10.1002/hep.28571
Page 25 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
1
Cirrhosis and Liver Cancer Mortality in the United States 1999-2016: An
Observational Study
Supplementary Material
Elliot B. Tapper MD (1,2) and Neehar D Parikh MD MS (1,2)
1. Division of Gastroenterology and Hepatology, University of Michigan
2. Gastroenterology Section, VA Ann Arbor Healthcare System, Ann Arbor
Table of Contents
Page 2: Methods - Technical supplement
Page 3: Supplementary Table 1a: Deaths due to Any Liver Disease Diagnosis (1999-2016)
Supplementary Table 1b: Deaths Involving Cirrhosis at Any Position on the Death
Certificate (1999-2016)
Supplementary Table 1C: Deaths Involving Both Cirrhosis and HCC (1999-2016)
Page 5: Supplementary Table 2b: The proportion of all deaths for each racial subgroup attributed to
cirrhosis Age >35
Supplementary Table 2b: The proportion of all deaths for each racial subgroup attributed to
cirrhosis Age >35
Page 6: Supplementary Table 3: Age-Adjusted Death Rates per 100,000 Americans
Page 7: Supplementary Table 4: Average Annual Percentage Change in the rate of Cirrhosis-related
death (2009-2016)
Page 9: Supplementary Table 5: Average Annual Percentage Change in the rate of HCC-related
death (2009-2016)
Page 11: Supplementary Table 6: Average Annual Percentage Change in the rate of Alcohol Use
Disorder-related Death (2009-2016)
Page 13: Supplementary Table 7: Overall Age-Adjusted Death Rates in States with High Rates of
Liver-Related Mortality
Page 15: Supplementary Table 8: Death-Rates Due to Other Conditions
Page 16: Supplementary Figure 1: Graphical Depiction of Cirrhosis Mortality
Page 26 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
2
Methods - Technical supplement
Database
CDC WONDER provides mortality data derived from US death certificates, compiled
from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative.
These include 50 states, 5 territories (Puerto Rico, U.S. Virgin Islands, Guam, American Samoa,
and the Northern Mariana Islands), the District of Columbia, and New York City. Each
jurisdiction reports data to National Center for Health Statistics though state centralized vital
record offices and the over 6,000 local vital registrars nationwide.
The CDC WONDER platform uses the number of cases as the numerator and the
corresponding population size based on US Census Bureau data as the denominator with death
rates calculated per 100,000 persons. Each estimate is associated with a standard error and 95%
confidence interval (CI), both based on the death count and the underlying population and
provided by the CDC WONDER platform. Within the platform, we used two specific data sets.
First, we used the Detailed Mortality File (DMF), a county-level national mortality and
population data collection derived from the U.S. records of deaths (death certificates) since 1968.
Counts and rates of death can be obtained by place of residence, age, race, gender, year, and
underlying cause-of-death. Second, we used the Multiple Cause of Death (MCD) data set which
contains the same mortality and population counts as does DMF. It is only available for 1999-
2015. MCD data contains a single underlying cause of death, and up to twenty additional coded
causes of death coupled with demographic data. This allowed us to assess trends in specific
causes of death linked to cirrhosis. Beginning in 1999 through 2015, all data available was coded
according to the International Classification of Diseases (ICD)-10 system. We therefore
restricted all analyses to this timeframe.
Page 27 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
3
Supplementary Table 1a: Deaths due to Any Liver Disease Diagnosis (1999-2016)
Supplementary Table 1b: Deaths Involving Cirrhosis at Any Position on the Death Certificate (1999-2016)
Liver Disease Deaths
Deaths Crude rate
(per 100,000) Age-adjusted (per 100,000)
Overall 742,800 17.11 17.07 17.15 16.23 16.19 16.26
Female Male
278,168 464,632
12.48 12.43 12.52 11.38 11.33 11.42
21.99 21.93 22.06 21.65 21.58 21.71
American Indian Asian/PI
African American White
16,257 11,435 70,542
644,562
32.02 31.53 32.51 37.84 37.23 38.45
5.03 4.93 5.12 6.23 6.11 6.35
12.77 12.68 12.87 13.99 13.89 14.1
18.36 18.31 18.4 16.79 16.75 16.84
Hispanic Non-Hispanic
95,141 645,165
15.91 15.81 16.01 23.36 23.2 23.51
17.23 17.19 17.27 15.56 15.52 15.59
Northeast Midwest South West
119,283 149,601 292,894 181,022
14.85 14.76 14.93 13.57 13.5 13.65
15.71 15.63 15.79 14.69 14.61 14.76
18.4 18.33 18.46 17.42 17.35 17.48
18.22 18.13 18.3 18.03 17.95 18.12
Liver Disease Deaths
Deaths Crude rate
(per 100,000) Age-adjusted (per 100,000)
Overall 800,344 18.4 18.4 18.5 17.4 17.3 17.4
Female Male
276,676 523,668
12.4 12.4 12.5 11.2 11.2 11.3
24.8 24.7 24.9 24.3 24.3 24.4
American Indian Asian/PI
African American White
14,383 15,892 80,485
689,584
28.3 27.9 28.8 34.1 33.5 34.6
7 6.9 7.1 8.8 8.6 8.9
14.6 14.5 14.7 15.9 15.8 16
19.6 19.6 19.7 17.8 17.8 17.9
Hispanic Non-Hispanic
111,480 685,901
18.6 18.5 18.8 27.9 27.7 28.1
18.3 18.3 18.4 16.4 16.4 16.4
Northeast Midwest South West
127,885 153,093 318,664 200,702
15.9 15.8 16 14.5 14.4 14.6
16.1 16 16.2 14.9 14.9 15
20 19.9 20.1 18.9 18.8 18.9
20.2 20.1 20.3 19.9 19.8 20
Page 28 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
4
Supplementary Table 1c: Deaths Involving Both Cirrhosis and HCC (1999-2016)
Cirrhosis and Hepatocellular Carcinoma
Deaths Crude rate
(per 100,000) Age-adjusted (per 100,000)
Overall 37737 0.56 0.53 0.59 0.56 0.53 0.59
Female Male
7136 30601
0.32 0.31 0.33 0.28 0.28 0.29
1.45 1.43 1.46 1.37 1.35 1.38
American Indian Asian/PI
African American White
511 2085 5169 29972
1.01 0.92 1.09 1.28 1.16 1.4
0.92 0.88 0.96 1.12 1.07 1.17
0.94 0.91 0.96 1.00 0.97 1.03
0.85 0.84 0.86 0.76 0.75 0.77
Hispanic Non-Hispanic
6522 31091
1.09 1.06 1.12 1.66 1.61 1.7
0.83 0.82 0.84 0.71 0.7 0.72
Northeast Midwest South West
6131 6773 13985 10848
0.76 0.74 0.78 0.68 0.67 0.7
0.71 0.69 0.73 0.63 0.62 0.65
0.88 0.86 0.89 0.79 0.78 0.81
1.09 1.07 1.11 1.04 1.02 1.06
Page 29 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
5
Supplementary Table 2a: The proportion of all deaths for each racial subgroup attributed to cirrhosis Age 25-34
Age 25-34
American Indian
Asian / Pacific
Islander African American White
Year Cirrhosis
Deaths
Proportion
of all
deaths
Cirrhosis
Deaths
Proportion
of all
deaths
Cirrhosis
Deaths
Proportion
of all
deaths
Cirrhosis
Deaths
Proportion
of all
deaths
2009 32 4.26% 8 0.75% 50 0.54% 294 0.93%
2010 33 4.15% 7 0.63% 54 0.61% 305 0.97%
2011 50 5.95% 8 0.71% 51 0.56% 337 1.03%
2012 39 4.18% 8 0.68% 44 0.48% 435 1.30%
2013 44 4.77% 17 1.38% 56 0.59% 426 1.26%
2014 51 5.41% 17 1.33% 48 0.50% 503 1.42%
2015 59 5.57% 18 1.30% 67 0.64% 568 1.47%
2016 76 6.26% 23 1.56% 64 0.54% 602 1.40%
Supplementary Table 2b: The proportion of all deaths for each racial subgroup attributed to cirrhosis Age >35
Age >35
American Indian
Asian / Pacific
Islander African American White
Year Cirrhosis
Deaths
Proportion
of all
deaths
Cirrhosis
Deaths
Proportion
of all
deaths
Cirrhosis
Deaths
Proportion
of all
deaths
Cirrhosis
Deaths
Proportion
of all
deaths
2009 753 5.81% 883 1.91% 4,278 1.65% 36,763 1.83%
2010 789 5.80% 909 1.89% 4,396 1.68% 38,672 1.90%
2011 860 6.16% 949 1.89% 4,570 1.73% 41,161 1.98%
2012 917 6.36% 1,064 2.00% 4,754 1.76% 43,188 2.06%
2013 989 6.56% 1,097 1.94% 5,235 1.89% 45,665 2.13%
2014 1,052 6.60% 1,164 2.00% 5,503 1.94% 48,305 2.24%
2015 1,194 7.10% 1,181 1.87% 5,695 1.95% 50,948 2.29%
2016 1,215 7.02% 1,266 1.93% 5,532 1.83% 51,426 2.30%
The denominator for ‘all deaths’ is national all cause mortality for the specified age group
Page 30 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
6
Supplementary Table 3: Age-Adjusted Death Rates per 100,000 Americans
Deaths attributable to cirrhosis
Deaths due to
HCC
(overall)
Year
Overall
Causes listed in addition to cirrhosis
Upper
Gastrointestinal
Bleeding
Peritonitis Sepsis Hepatorenal
syndrome
Cerebro-
vascular and
ischemic
heart disease
Trauma
1999 9.58 9.45 9.71 1.63 1.58 1.68 0.29 0.26 0.31 0.72 0.68 0.76 0.91 0.87 0.96 0.49 0.46 0.52 0.21 0.19 0.23 2.85 2.77 2.93
2000 9.62 9.49 9.75 1.61 1.56 1.67 0.31 0.28 0.34 0.72 0.68 0.76 0.83 0.79 0.87 0.52 0.49 0.55 0.23 0.2 0.25 3 2.92 3.08
2001 9.58 9.45 9.71 1.59 1.53 1.64 0.28 0.26 0.3 0.73 0.7 0.77 0.84 0.8 0.88 0.49 0.47 0.52 0.2 0.18 0.22 2.95 2.87 3.03
2002 9.61 9.48 9.74 1.46 1.41 1.51 0.31 0.29 0.34 0.7 0.67 0.74 0.82 0.78 0.86 0.54 0.51 0.57 0.2 0.18 0.22 3.06 2.98 3.14
2003 9.7 9.58 9.83 1.47 1.42 1.52 0.28 0.26 0.31 0.72 0.69 0.76 0.8 0.76 0.84 0.5 0.47 0.53 0.21 0.19 0.23 3.21 3.13 3.29
2004 9.39 9.26 9.51 1.39 1.34 1.44 0.27 0.25 0.29 0.76 0.72 0.79 0.77 0.73 0.81 0.44 0.41 0.47 0.18 0.16 0.2 3.21 3.13 3.29
2005 9.42 9.3 9.54 1.25 1.21 1.3 0.27 0.24 0.29 0.79 0.76 0.83 0.71 0.67 0.75 0.47 0.44 0.5 0.21 0.19 0.23 3.27 3.19 3.35
2006 9.19 9.07 9.31 1.16 1.12 1.21 0.25 0.23 0.28 0.75 0.71 0.78 0.66 0.62 0.69 0.47 0.44 0.5 0.18 0.16 0.2 3.34 3.26 3.42
2007 9.28 9.16 9.4 1.23 1.19 1.27 0.23 0.21 0.25 0.77 0.73 0.8 0.7 0.66 0.74 0.46 0.43 0.48 0.2 0.18 0.22 3.32 3.24 3.4
2008 9.36 9.24 9.48 1.25 1.21 1.29 0.25 0.23 0.27 0.79 0.75 0.82 0.68 0.64 0.72 0.47 0.44 0.5 0.19 0.17 0.21 3.54 3.46 3.62
2009 9.36 9.25 9.48 1.25 1.21 1.3 0.28 0.25 0.3 0.82 0.78 0.85 0.64 0.61 0.68 0.44 0.42 0.47 0.18 0.16 0.2 3.66 3.58 3.75
2010 9.72 9.6 9.84 1.29 1.25 1.34 0.28 0.26 0.3 0.9 0.87 0.94 0.69 0.65 0.72 0.49 0.47 0.52 0.17 0.15 0.19 3.76 3.68 3.84
2011 10.1 9.98 10.23 1.31 1.27 1.36 0.27 0.25 0.29 0.91 0.88 0.95 0.71 0.67 0.74 0.48 0.46 0.51 0.21 0.19 0.23 3.87 3.79 3.96
2012 10.38 10.26 10.5 1.36 1.32 1.41 0.3 0.28 0.32 0.99 0.96 1.03 0.72 0.69 0.76 0.51 0.49 0.54 0.19 0.18 0.21 3.83 3.75 3.91
2013 10.65 10.53 10.77 1.43 1.39 1.48 0.31 0.28 0.33 1.03 0.99 1.06 0.72 0.68 0.75 0.52 0.49 0.55 0.18 0.17 0.2 3.9 3.82 3.98
2014 11.09 10.97 11.21 1.38 1.34 1.43 0.36 0.34 0.39 1.13 1.1 1.17 0.76 0.72 0.79 0.5 0.47 0.53 0.19 0.17 0.2 3.84 3.76 3.92
2015 11.57 11.45 11.7 1.46 1.42 1.51 0.4 0.37 0.43 1.25 1.2 1.29 0.83 0.79 0.86 0.56 0.54 0.59 0.19 0.17 0.21 3.93 3.85 4.01
2016 11.42 11.29 11.54 1.56 1.51 1.6 0.37 0.35 0.4 1.29 1.24 1.33 0.85 0.81 0.88 0.55 0.52 0.58 0.23 0.21 0.25 4.2 4.12 4.28
AAPC 1.1 (0.8-1.4) -0.5 (-1.8- -0.9) 1.5 (0.2-2.8) 3.7 (3.2-4.2) -0.6 (-1.2 – 0.1) 0.6 (-0.3 – 1.5) -0.2 (-1.1-0.6) 2.1 (1.3-3.0)
AAPC
Since
2009 3.0 (2.5-3.4) 2.6 (2.0-3.2) 6.1 (3.9-8.2) 7.1 (6.1-8.4) 3.2 (2.2 – 4.2) 2.9 (1.5-4.4) -0.2 (-1.1-0.6) 2.0 (0.2-3.7)
AAPC = Average annual percentage change. All percentage changes are presented with 95% Confidence Intervals. AAPC Intervals that do not
cross zero have p values < 0.05. Gastrointestinal bleeding is defined by the presence on the death certificate of causes inclusive of peptic ulcers,
Page 31 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
7
melena, hematemesis and bleeding varices. Peritonitis and hepatorenal syndrome are both defined by the corresponding ICD-10 codes. Trauma is
defined by V01-Y89.
Page 32 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
8
Supplementary Table 4: Average Annual Percentage Change in the rate of Cirrhosis-related
death (2009-2016)
State AAPC 95% CI Age-Adjusted Death-Rate in
2009 (95% CI)
Alabama 5 3.2 6.8 11.11 10.06 12.15
Alaska 2.3 0.7 3.9 11.18 8.19 14.91
Arizona 3.6 2.4 4.9 12.7 11.7 13.7
Arkansas 5.7 3.9 7.6 9.6 8.35 10.84
California 2.7 2.1 3.2 11.77 11.36 12.18
Colorado 2.4 2 2.9 10.52 9.48 11.56
Connecticut 3.3 1.5 5.2 8.17 7.15 9.18
Delaware 0 -1.4 1.4 10.47 8.27 13.06
District of Columbia -1.7 -3.5 0.1 9.07 6.42 12.45
Florida 2.2 1 3.3 12.19 11.65 12.72
Georgia 3.9 1.4 6.5 8.47 7.78 9.16
Hawaii 2.3 1.3 3.4 7.39 5.9 9.16
Idaho 2.4 1.6 3.3 8.6 6.91 10.29
Illinois 2.1 1.1 3.2 9.05 8.46 9.64
Indiana 5 3.8 6.1 9.08 8.26 9.91
Iowa 3.1 2.1 4.2 7.22 6.18 8.27
Kansas 1.7 1 2.3 7.77 6.61 8.93
Kentucky 6.8 5.1 8.5 9.9 8.86 10.94
Louisiana 4.5 2.2 6.9 9.19 8.19 10.2
Maine 1.4 0.5 2.2 9.2 7.51 10.9
Maryland -1.2 -1.7 -0.7 8.09 7.26 8.92
Massachusetts -0.3 -0.9 0.2 8.78 7.99 9.56
Michigan 1.6 1 2.1 10.47 9.77 11.17
Minnesota 3.3 2.2 4.4 6.93 6.14 7.73
Mississippi 3.9 2.4 5.5 11.07 9.71 12.43
Missouri 3.6 2.6 4.6 8.3 7.5 9.11
Page 33 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
9
Montana 2.2 1.3 3.1 11.25 9.09 13.77
Nebraska 1.6 0.5 2.7 6.71 5.44 8.19
Nevada 0.7 -0.3 1.7 11.82 10.33 13.32
New Hampshire 1.6 0.3 2.9 9.57 7.78 11.37
New Jersey 1 0.2 1.9 8.3 7.63 8.96
New Mexico 6 4.1 7.9 16.99 14.97 19
New York 1.6 0.9 2.3 6.92 6.51 7.33
North Carolina 2.8 1.9 3.7 10.44 9.7 11.17
North Dakota 5.9 -0.7 12.8 5.7 3.84 8.13
Ohio 3.3 2.8 3.8 10.2 9.57 10.84
Oklahoma 2.6 1.8 3.5 11.4 10.18 12.63
Oregon 2.6 2.1 3 10.68 9.54 11.82
Pennsylvania 2.1 1.5 2.7 8.91 8.35 9.46
Rhode Island 0.5 -0.6 1.7 10.43 8.41 12.8
South Carolina 1.9 1.3 2.6 12.27 11.15 13.4
South Dakota 2.7 1.1 4.3 6.74 4.9 9.04
Tennessee 3.8 3 4.7 11.68 10.74 12.61
Texas 3.2 2.3 4.1 14.09 13.53 14.65
Utah 4 1.9 6.2 6.97 5.66 8.28
Vermont 1.1 -0.4 2.7 8.8 6.53 11.61
Virginia 3.2 2 4.4 9.1 8.35 9.84
Washington 2.2 1.5 3 9.83 8.99 10.68
West Virginia 2.1 1.1 3.1 12.54 10.85 14.24
Wisconsin 1.5 0.9 2.1 8.19 7.37 9.01
Wyoming 3.1 1.2 5 7.55 5.23 10.56
Page 34 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
10
Supplementary Table 5: Average Annual Percentage Change in the rate of HCC-related
death (2009-2016)
State AAPC 95% CI Age Adjusted Death-Rate in
2009 (95% CI)
Alabama -1.1 -4.4 2.3 2.93 2.39 3.47
Arizona 5.3 -4.7 16.3 2.7 2.24 3.16
Arkansas 10.9 3.5 18.9 1.84 1.34 2.47
California 0.8 0.4 1.2 4.28 4.03 4.53
Colorado 3.8 2.5 5 3.32 2.72 3.92
Connecticut 1.7 0.3 3 3.11 2.52 3.8
Delaware 2.3 0.6 4 3.47 2.26 5.08
District of Columbia 2.3 -0.6 5.3 7.0 4.72 9.99
Florida 1.1 -3.1 5.5 3.29 3.02 3.57
Georgia 2.5 1.8 3.2 2.38 2.01 2.75
Hawaii 0 -1.1 1.2 6.27 4.91 7.88
Idaho 5.5 2 9.2 1.81 1.12 2.76
Illinois 1.1 0.4 1.7 2.72 2.4 3.05
Indiana 3.1 -0.4 6.6 2.7 2.25 3.15
Iowa 4.5 3 6 2.48 1.91 3.17
Kansas 5.2 3.8 6.7 2.73 2.09 3.5
Kentucky 4.2 3.2 5.1 3.04 2.46 3.62
Louisiana 2.5 1.4 3.6 3.19 2.6 3.79
Maine 2.4 0.6 4.3 3.12 2.22 4.27
Maryland 2.6 1.9 3.4 3.18 2.66 3.71
Massachusetts -0.7 -2.8 1.5 3.87 3.34 4.39
Michigan 2.4 1.4 3.4 2.58 2.24 2.93
Minnesota 3.1 2 4.2 2.63 2.14 3.12
Mississippi 2.5 0.6 4.4 1.69 1.2 2.31
Missouri 0.2 -2 2.4 3.32 2.81 3.83
Nebraska 3.9 1.5 6.3 3.07 2.24 4.1
Page 35 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
11
Nevada 3.2 2 4.4 2.77 2.08 3.62
New Hampshire 2.4 0.8 3.9 2.35 1.53 3.44
New Jersey 0.3 -0.5 1 2.83 2.44 3.22
New Mexico 2.5 0.7 4.3 3.44 2.61 4.46
New York 1.3 0.8 1.9 3.01 2.74 3.28
North Carolina 4.2 3.4 5.1 3.7 3.26 4.14
Ohio 1.5 -0.2 3.2 3.2 2.85 3.56
Oklahoma 7.4 -1.1 16.6 2.83 2.25 3.5
Oregon 5.2 3.9 6.5 4.75 4 5.51
Pennsylvania 2.9 2.3 3.6 3.37 3.03 3.71
Rhode Island 1.2 -0.6 3.1 3.84 2.66 5.37
South Carolina 3.4 2.4 4.5 3.78 3.16 4.4
Tennessee 4.1 3.1 5.1 3.98 3.43 4.52
Texas 2.3 1.6 2.9 4.24 3.92 4.55
Utah 2.5 0.8 4.2 2.23 1.55 3.1
Vermont -5.6 -12.2 1.5 5.19 3.48 7.46
Virginia 0.5 -4.2 5.5 3.77 3.29 4.26
Washington 1.6 -1.1 4.5 4.51 3.93 5.09
West Virginia 3.1 1.1 5.1 2.05 1.44 2.84
Wisconsin 2.8 2.1 3.4 2.4 1.96 2.84
Page 36 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
12
Supplementary Table 6: Average Annual Percentage Change in the rate of Alcohol Use
Disorder-related Death (2009-2016)
State AAPC 95% CI Age Adjusted Rate in 2006
Alabama -2.3 -3.6 -1.1 1.59 1.23 2.03
Alaska -0.8 -9.5 8.7 8.89 6.26 12.25
Arizona 0.9 0 1.8 3.71 3.18 4.24
Arkansas -0.5 -2.4 1.3 1.54 1.09 2.11
California 0.7 -1 2.5 2.25 2.08 2.42
Colorado 0.4 -0.8 1.5 4.85 4.17 5.53
Connecticut 7.8 -0.5 16.9 1.79 1.35 2.33
Delaware 11.1 -7.2 33 2.75 1.7 4.2
District of Columbia -6.4 -8.5 -4.3 4.17 2.55 6.44
Florida 5.9 3.8 8 2.45 2.21 2.69
Georgia 2.3 0.8 3.9 2.68 2.31 3.04
Idaho 0.9 -1.1 2.9 2.88 2.01 4.01
Illinois 8 4.6 11.6 1.72 1.47 1.98
Indiana 8.3 4.8 11.8 2.04 1.66 2.42
Iowa 13.1 8.9 17.4 1.8 1.31 2.4
Kansas 2.9 -2.8 8.9 2.87 2.23 3.63
Kentucky 7.6 2.4 13 2.83 2.29 3.37
Louisiana -1.6 -3.5 0.4 1.37 1.01 1.81
Maine 10.8 5.8 16 2.04 1.34 3
Maryland -2.5 -3.8 -1.2 1.8 1.45 2.22
Massachusetts 7.3 4.5 10.1 2.22 1.83 2.61
Michigan -0.2 -1.3 0.9 2.6 2.26 2.95
Minnesota 4.8 -1.6 11.6 2.42 1.97 2.87
Mississippi -2.8 -4.3 -1.3 2.4 1.83 3.09
Missouri 5.7 -1.1 12.9 2.19 1.78 2.59
Montana 3.4 1.8 5.1 5.46 3.97 7.34
Nebraska 15.4 9.2 22 - 0.66 1.82
Page 37 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
13
Nevada 0.5 -0.8 1.8 4.79 3.88 5.71
New Hampshire 2.1 0.6 3.6 3.1 2.19 4.28
New Jersey -1.8 -3.4 -0.2 1.22 0.99 1.49
New Mexico 7.3 3 11.8 8.25 6.84 9.65
New York 3.3 0.4 6.3 2.06 1.85 2.28
North Carolina 5.1 2.1 8.2 2.29 1.95 2.62
North Dakota -3.7 -12.3 5.8 - 1.94 5.33
Ohio 10.2 6.1 14.5 1.37 1.14 1.6
Oklahoma 4.3 2.3 6.2 2.4 1.88 3.01
Oregon 6.8 2.2 11.5 4.4 3.68 5.12
Pennsylvania 3.9 2.5 5.2 1.63 1.39 1.88
Rhode Island 7.8 0.7 15.4 1.99 1.2 3.1
South Carolina 4.1 -4.2 13.1 2.39 1.92 2.87
South Dakota 3.2 -9.3 17.4 3.91 2.53 5.77
Tennessee 7.3 3.7 11 2.16 1.77 2.55
Texas -0.8 -1.8 0.3 1.66 1.47 1.84
Utah 0.8 -0.5 2.2 3.1 2.32 4.05
Vermont 8.5 -1.3 19.3 - 1.45 4.27
Virginia 5.8 2.6 9.1 2.01 1.67 2.35
Washington 0.9 -0.4 2.2 2.12 1.73 2.51
West Virginia 12.3 4.7 20.5 1.85 1.27 2.6
Wisconsin 5.1 0.7 9.7 2.87 2.4 3.34
Wyoming 10.6 -0.8 23.4 - 2.03 5.78
Page 38 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
14
Supplementary Table 7: Overall Age-Adjusted Death Rates in States with High Rates of Liver-Related Mortality
Alabama Arizona California Kansas Louisiana Maryland New Mexico
Oregon Texas
1999 1575.13 1284.53 1056.49 1336.62 1337.65 1420.2 1310.48 1322.71 1397.74
2000 1572.77 1268.36 1036.79 1338.39 1320.42 1426.56 1255.16 1311.03 1404.93
2001 1570.87 1243.73 1031.74 1326.52 1326.40 1390.96 1297.52 1314.37 1403.18
2002 1587.94 1269.77 1013.31 1336.7 1317.31 1365.93 1282.98 1325.37 1399.6
2003 1602.18 1248.12 1011.31 1306.71 1328.39 1350.22 1296.66 1289.92 1361.31
2004 1563.01 1205.73 966.22 1252.34 1290.91 1285.56 1223.48 1237.8 1316.14
2005 1570.39 1230.14 961.39 1283.9 1339.91 1282.8 1245.50 1242.79 1309.44
2006 1529.44 1196.42 945.22 1260.32 1251.52 1243.26 1236.37 1218.78 1271.30
2007 1494.87 1140.84 910.00 1235.33 1224.84 1223.52 1211.42 1193.24 1264.45
2008 1505.97 1121.34 893.43 1247.24 1239.18 1203.17 1228.56 1189.28 1267.14
2009 1478.56 1089.00 863.54 1191.93 1187.98 1178.37 1168.50 1155.11 1217.51
2010 1476.99 1092.21 854.09 1201.38 1186.41 1146.33 1170.96 1147.54 1220.59
2011 1469.05 1085.20 846.92 1210.45 1163.76 1125.39 1172.05 1146.91 1189.67
2012 1458.72 1076.35 833.93 1198.99 1179.60 1116.41 1162.72 1120.65 1192.09
2013 1456.80 1064.19 833.05 1195.96 1178.60 1119.41 1150.01 1139.97 1189.38
2014 1430.24 1043.43 801.77 1200.42 1175.71 1101.90 1172.83 1122.19 1179.28
2015 1455.28 1061.53 823.11 1225.52 1148.29 1108.99 1156.97 1148.22 1178.86
2016 1442.28 1067.18 817.04 1195.24 1144.04 1125.17 1177.12 1122.11 1155.76
AAPC 2009-2016
-0.4 (-0.7 – 0.01)
-0.5 (-1.0 – 0.0)
-1.0 (-1.2 - -0.9)
-0.1 (-0.6–0.4)
-1.0 (-1.5 – -0.6)
-0.7 (-1.2 - -0.3)
-0.7 (-0.9 - -0.6)
-0.5 (-1.1 – 0.2)
-0.8 (-1.3 - -0.3)
AAPC = average annual percentage change
Page 39 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
15
Supplementary Table 8: Death-Rates Due to Other Conditions
Cirrhosis Infections Neoplasia Cardiovascular Disease
Respiratory Disease
Age-adjusted Rate
Total Deaths
Age-adjusted Rate
Total Deaths
Age-adjusted Rate
Total Deaths
Age-adjusted Rate
Total Deaths
Age-adjusted Rate
Total Deaths
1999 9.58 20,661 32.94 58,447 301.72 534,316 540.97 951,686 129.37 227,985
2000 9.62 21,003 32.1 57,611 299.75 537,000 526.41 938,843 128.52 229,285
2001 9.58 21,390 32.22 58,805 294.82 537,278 509.62 924,331 125.9 228,203
2002 9.61 21,835 33.8 62,526 291.3 540,015 500.17 920,716 126.67 232,826
2003 9.7 22,496 33.68 63,291 285.75 539,069 482.85 904,465 125.19 233,938
2004 9.39 22,166 33.18 63,312 279.24 535,440 453.9 863,280 118.89 225,342
2005 9.42 22,808 33.53 65,151 276.45 540,134 441.79 858,251 123.15 237,950
2006 9.19 22,717 33.1 65,664 271.39 540,415 415.53 825,146 114.19 224,931
2007 9.28 23,473 32.42 65,578 267.38 542,761 397.93 807,609 112.23 225,547
2008 9.36 24,140 32.55 67,222 262.61 544,306 388.65 805,912 118.66 243,365
2009 9.36 24,736 32.32 68,090 257.88 545,370 369.49 782,260 113.08 236,444
2010 9.72 26,045 30.87 66,178 256.54 551,580 363.02 782,081 110.96 235,777
2011 10.1 27,614 30.67 67,575 250.33 552,251 352.3 781,117 112.94 246,951
2012 10.38 28,971 29.84 67,193 246.12 556,760 345.67 785,154 109.18 244,817
2013 10.65 30,293 30.16 69,444 240.96 558,122 343.61 798,743 112.75 259,381
2014 11.09 32,142 29.33 69,208 237.85 564,249 338.93 805,554 108.96 256,642
2015 11.57 33,968 29.43 70,856 233.52 567,484 343.32 834,308 111.64 269,543
2016 11.42 34,174 28.07 68,791 229.16 568,787 338.26 838,482 107.23 264,353
Average
Annual
Percentage
Change
1999-2016
1.1 (0.8 – 1.4) -0.8 (-1.1 - -0.4) -1.6 (-1.7 - -1.6) -2.7 (-3.2 - -2.3) -1.1 (-1.3 – -0.8)
The absolute number of deaths and age-adjusted death-rate (per 100,000 people) is shown for each year 1999-2016 for cirrhosis, infections (ICD-
10 A00-B99), neoplasia (C00-C48, not including HCC, C22), cardiovascular disease (I00-I99), and Respiratory Disease (J00-J98). The
average annual percentage change is calculated for the whole study period and shown as percentage (95% confidence interval).
Page 40 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
16
Supplementary Figure 1: Graphical Depiction of Cirrhosis Mortality
Page 41 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
Figure 1: Death-rate Trends by Age-Group 2009-2016 This graph explores changes in mortality by subgroup over the period in which there was an overall increase in cirrhosis mortality. Each bar represents the average annual percentage change in the death-rate due to
cirrhosis and hepatocellular carcinoma (HCC) in all adult age groups from 2009-2016. Panel A depicts changes in overall cirrhosis and HCC mortality; B shows changes by race; C shows changes by sex; and D
shows changes by alcohol-related diagnosis codes.
176x114mm (300 x 300 DPI)
Page 42 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
Figure 2: Age Adjusted Death-rates Attributable to Cirrhosis and Hepatocellular Carcinoma over Time The age-adjusted mortality for each condition in each state is presented in quartiles for the first, middle and last year of the study. Quartiles for each time period are denoted by yellow (first), light orange (second), dark orange (third), and red (fourth). Striped or shaded states imply data that is either unreliable or
suppressed to protect patient identity.
189x129mm (300 x 300 DPI)
Page 43 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
Figure 3: States with Significant changes in Death-rates Attributable to Cirrhosis and Hepatocellular Carcinoma (HCC) 2009-2016
This graph explores the changes in cirrhosis and HCC mortality at the state level 2009-2016. The average annual percentage change (AAPC) in mortality for each individual state is plotted against the age-adjusted death-rate in 2009. States without significant trends were excluded from this figure for simplicity. Panel A
shows trends for cirrhosis and B shows trends for HCC.
220x373mm (300 x 300 DPI)
Page 44 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
Figure 4: Average Annual Percentage Change (AAPC) in Mortality Due to Alcohol-Related Disease Note: Many states were excluded from the figures for simplicity because their trends were non-significant
(no changes over the specified timeframes).
Top: The AAPC for mortality by state for alcohol-related cirrhosis (ICD-10 K70.3) is shown for 2009-2016. Bottom: The AAPC for mortality by state for alcohol-use disorder (ICD-10 F10) is shown for 2009-2016.
259x342mm (300 x 300 DPI)
Page 45 of 45
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960