Vivarium Research
  • Vivarium Model Design
  • Vivarium Models
    • Simulation Concept Models
    • Intervention Models
    • Cause Models
      • Atrial fibrillation and flutter
      • Alzheimer’s Disease
      • Aortic Aneurysm
      • Archived cause models
      • Chronic Obstructive Pulmonary Disease
      • Diabetes Mellitus (PAF of 1 with FPG)
      • Diarrheal diseases
      • Hypertensive Heart Disease (PAF of 1 with hypertension)
      • Intracerebral Hemorrhage
      • Ischemic Heart Disease
      • Ischemic Stroke
      • Latent Tuberculosis
      • Lower Respiratory Infections
      • Malaria: GBD 2021
      • Maternal disorders
      • Maternal hemorrhage incidence
      • Measles
      • Neonatal all-cause mortality: GBD 2021, MNCNH
      • Neoplasm (Cancer) Cause Models
      • Neural Tube Defects
      • Background morbidity for the nutrition optimization pregnancy model
      • Peripheral Arterial Disease
      • Postpartum depression
      • Smoking Related Mortality (Excluding Lung Cancer)
        • GBD 2017 Modeling Strategy
        • Vivarium Modeling Strategy
        • References
      • Subarachnoid Hemorrhage
    • Risk Exposure Models
    • Risk-Attributable Cause Models
    • Risk Effects Models
    • Risk Correlation Models
    • Other Models
    • Vivarium Model Document Templates
  • Onboarding Resources
  • Glossary
Vivarium Research
  • Vivarium Models
  • Cause Models
  • Smoking Related Mortality (Excluding Lung Cancer)
  • View page source

Smoking Related Mortality (Excluding Lung Cancer)

Mortality among smokers is much higher than mortality among non-smokers. Therefore, it can be important to consider increased mortality attributable to smoking when smoking is included in a model, even if specific causes affected by smoking are not necessary to explicitly include in the model. See the causal framework section in the Vivarium CSU Lung Cancer Screening Concept Model Documentation for an example of when this is important.

The purpose of this cause model document is to represent the increased mortality risk associated with smoking exposure by combining the causes affected by the smoking risk factor. This specific cause model document was created to exclude lung cancer so that it can be used in a model that models the lung cancer cause directly.

  • GBD 2017 Modeling Strategy

  • Vivarium Modeling Strategy

  • References

Abbreviations

Abbreviation

Definition

Note

SACs

Smoking-affected causes

ACMR

All-cause mortality rate

CSMR

Cause-specific mortality rate

EMR

Excess mortality rate

RF

Risk factor

GBD 2017 Modeling Strategy

The smoking risk factor in GBD 2017 has the following outcome pairs:

tuberculosis, lower respiratory tract infections, oesophageal cancer, stomach cancer, bladder cancer, liver cancer, laryngeal cancer, lung cancer, breast cancer, cervical cancer, colorectal cancer, lip and oral cancer, nasopharyngeal cancer, other pharyngeal cancer, pancreatic cancer, kidney cancer, leukaemia, ischaemic heart disease, ischaemic stroke, haemorrhagic stroke, subarachnoid haemorrhage, atrial fibrillation and flutter, aortic aneurysm, peripheral arterial disease, chronic obstructive pulmonary disease, other chronic respiratory diseases, asthma, peptic ulcer disease, gallbladder and biliary tract diseases, Alzheimer disease and other dementias, Parkinson disease (protective), multiple sclerosis, type‐II diabetes, rheumatoid arthritis, low back pain, cataracts, macular degeneration, and fracture.

[GBD-2017-CoD-Appendix-Smoking], [GBD-2017-YLD-Appendix-Smoking]

Note

These risk-outcome pairs are consistent between GBD 2017 and GBD 2019

Cause Hierarchy

Smoking-Affected Cause IDs

Cause

Cause ID

Age group ID start

Restrictions

More detailed subcauses?

Notes

tuberculosis

297

4

True; Drug susceptible/resistant subcauses

lower respiratory tract infections

322

2

oesophageal cancer

411

8

stomach cancer

414

8

bladder cancer

474

8

liver cancer

417

8

True; Etiology sub causes

laryngeal cancer

423

8

lung cancer

426

8

Will be excluded for this cause model

breast cancer

429

8

cervical cancer

432

8

FEMALE ONLY

colorectal cancer

441

8

lip and oral cancer

444

8

nasopharyngeal cancer

447

6

other pharyngeal cancer

450

8

pancreatic cancer

456

8

kidney cancer

471

2

leukaemia

487

2

True; Subtype subcauses

ischaemic heart disease

493

8

ischaemic stroke

495

2

haemorrhagic stroke (Intracerebral hemorrhage)

496

2

subarachnoid haemorrhage

497

2

atrial fibrillation and flutter

500

11

Age group start after lung cancer

aortic aneurysm

501

8

peripheral arterial disease

502

13

Age group start after lung cancer

chronic obstructive pulmonary disease

509

4

other chronic respiratory diseases

520

2

asthma

515

5

peptic ulcer disease

527

5

gallbladder and biliary tract diseases

534

5

Alzheimer disease and other dementias

543

13

Age group start after lung cancer

Parkinson disease (protective)

544

9

Age group start after lung cancer

multiple sclerosis

546

9

Age group start after lung cancer

type‐II diabetes

976

8

rheumatoid arthritis

627

6

low back pain

630

N/A

YLD only

Exclude due to nonfatal

cataracts

671

N/A

YLD only

Exclude due to nonfatal

macular degeneration

672

N/A

YLD only

Exclude due to nonfatal

Fracture

N/A

N/A

YLD only

Exclude due to nonfatal

Note

The starting age group IDs were extracted according to the YLL start, not the YLD start.

The Vivarium CSU Lung Cancer Screening Model only models simulants 15 years (age group ID 8) and older, so notes were made for causes with age group starts older than 15 years only.

Restrictions

See the table above.

Vivarium Modeling Strategy

Scope

The Vivarium modeling strategy outlined in this document will be a mortality only model that utilizes forecasted cause-specific mortality rates for ischemic heart disease (IHD) and chronic obstructive pulmomary disease (COPD) from 2020 to 2040. These causes were selected because they represent a large share of all smoking-related mortality without needing to model several additional causes with smaller contributions. This model is intended to be pair with the smoking risk exposure model.

This model was designed for use in the Lung Cancer Screening model, which uses data forecasted from 2020 to 2040.

Assumptions and Limitations

This model is limited in that it only considers smoking attributable deaths due to IHD and COPD (as well as lung cancer when modeled as a part of the Lung Cancer Screening model) and not the totality of smoking attributable deaths.

According to GBD 2019 estimates, lung cancer, COPD, and IHD combined account for 65.6% (UI: 62.6 - 67.9) of all smoking-attributable mortaltiy in China across all age groups and both sexes. For males, the value is 64.8% and for females it is 71.1%. The calculation of these values is hosted here.

Note

If stroke (cause ID 494) were to be added to this mortality model, which would require forecasted mortality estimates, 82.1% of smoking attributable mortality among the Chinese population would be represented in our model. This was not included in the current modeling strategy due to scope constraints on the forecast models.

Cause Model Diagram

There is no cause model diagram for this cause model because it is a mortality-only model. Mortality using should be modeled as follows:

\[ \begin{align}\begin{aligned}mr_i = ACMR - CSMR_\text{426} - (\sum_{c=1}^{n} CSMR_c)\\+ EMR(i)_\text{426} * (1 - PAF_\text{426}) * RR(i)_\text{426}\\+ (\sum_{c=1}^{n} CSMR_c * (1 - PAF_c) * RR(i)_c)\end{aligned}\end{align} \]

Where,

Definitions

Parameter

Definition

Note

\(mr_i\)

Mortality rate for an individual simulant

\(ACMR\)

All cause mortality rate forecasted from 2020-2040

/ihme/csu/swiss_re/forecast

\(426\)

Lung cancer

\(c\)

509, 493, 414, 441

COPD, IHD, Stomach cancer, Colorectal cancer

\(CSMR_\text{426}\)

Cause-specific mortality rate for lung cancer forecasted from 2020-2040

Defined in the lung cancer cause model document

\(CSMR_c\)

Cause-specific mortality rate for cause \(c\), forecasted from 2020-2040

/ihme/csu/swiss_re/forecast/{c}_deaths_12_29_ng_smooth_13.nc, ‘nosed_forecast’ column

\(RR(i)_\text{c}\)

Relative risk of a given cause for an individual simulant based on their smoking exposure

Defined in the smoking risk effects documentation page

\(PAF_\text{c}\)

Population attributable fraction for smoking on mortality due to cause c

\(\frac{\overline{RR(i)_c} - 1}{\overline{RR(i)_c}}\) (details in the smoking risk effects page)

\(EMR(i)_\text{426}\)

Excess mortality rate of lung cancer for an individual simulant (based on cause model state)

Defined in the lung cancer cause model document; use forecasted 2019 value

Note

May also include cause 494 (stroke) in cause list \(c\) if/when it is added to the forecast models

Upon the event that a simulant dies according to their \(mr_i\), the probability that their cause of death was one of the following causes is represented in the table below:

Cause of death likelihoods

Cause of death

Probability

Note

Lung cancer

\(\frac{EMR(i)_\text{426} * (1 - PAF_\text{426}) * RR(i)_\text{426}}{mr_i}\)

Smoking-affected causes

\(\frac{\sum_{c=1}^{n} CSMR_c * (1 - PAF_c) * RR(i)_c}{mr_i}\)

For \(c\) in [493,509,414,441]

Other causes

\(\frac{1 - EMR(i)_\text{426} * (1 - PAF_\text{426}) * RR(i)_\text{426} + (\sum_{c=1}^{n} CSMR_c * (1 - PAF_c) * RR(i)_c)}{mr_i}\)

Validation Criteria

The simulation output should replicate the cause-specific mortality rate of smoking affected causes as defined in the data tables. Additionally, GBD all cause mortality should be replicated in the simulation output.

When simulation output is stratified on smoking status, never smokers should have lower mortality rates than current and former smokers.

References

[GBD-2017-YLD-Appendix-Smoking]

Pages ???-??? in Supplementary appendix 1 to the GBD 2017 YLD Capstone:

(GBD 2017 YLD Capstone) GBD 2017 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018; 392: 1789–858. DOI: https://doi.org/10.1016/S0140-6736(18)32279-7

[GBD-2017-CoD-Appendix-Smoking]

Pages ???-??? in Supplementary appendix 1 to the GBD 2017 CoD Capstone:

(GBD 2017 CoD Capstone) GBD 2017 Causes of Death Collaborators. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 2018; 392: 1736–88. DOI: http://dx.doi.org/10.1016/S0140-6736(18)32203-7

Previous Next

© Copyright 2025, The Vivarium Research Team.

Built with Sphinx using a theme provided by Read the Docs.