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.
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
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:
Where,
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 |
|
\(RR(i)_\text{c}\) |
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 |
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
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
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