Projects / Programmes
Years of life lost as a measure of disease burden
Code |
Science |
Field |
Subfield |
3.08.00 |
Medical sciences |
Public health (occupational safety) |
|
Code |
Science |
Field |
P160 |
Natural sciences and mathematics |
Statistics, operations research, programming, actuarial mathematics |
Code |
Science |
Field |
3.03 |
Medical and Health Sciences |
Health sciences |
Survival analysis, relative survival, years lost, pseudo-observations, competing risks
Researchers (15)
no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
1. |
30722 |
PhD Rok Blagus |
Systems and cybernetics |
Researcher |
2019 - 2020 |
205 |
2. |
39138 |
Rok Hribar |
Computer science and informatics |
Technical associate |
2022 - 2023 |
24 |
3. |
24344 |
PhD Nataša Kejžar |
Systems and cybernetics |
Researcher |
2019 - 2022 |
158 |
4. |
56197 |
Tina Košuta |
Medical sciences |
Researcher |
2021 - 2023 |
10 |
5. |
56836 |
Maša Kušar |
Medical sciences |
Researcher |
2022 - 2023 |
8 |
6. |
15355 |
PhD Branimir Leskošek |
Public health (occupational safety) |
Researcher |
2019 - 2023 |
184 |
7. |
29917 |
PhD Lara Lusa |
Public health (occupational safety) |
Researcher |
2022 - 2023 |
253 |
8. |
51959 |
PhD Damjan Manevski |
Public health (occupational safety) |
Researcher |
2019 - 2023 |
45 |
9. |
57119 |
Urh Peček |
|
Technical associate |
2022 - 2023 |
0 |
10. |
51964 |
PhD Jakob Peterlin |
Mathematics |
Researcher |
2022 - 2023 |
12 |
11. |
23437 |
PhD Maja Pohar Perme |
Public health (occupational safety) |
Head |
2019 - 2023 |
297 |
12. |
33230 |
PhD Nina Ružić Gorenjec |
Mathematics |
Researcher |
2019 - 2023 |
55 |
13. |
08992 |
PhD Janez Stare |
Public health (occupational safety) |
Researcher |
2019 - 2023 |
280 |
14. |
36368 |
PhD Marko Vidak |
Medical sciences |
Researcher |
2022 - 2023 |
25 |
15. |
17837 |
PhD Gaj Vidmar |
Systems and cybernetics |
Researcher |
2019 - 2022 |
561 |
Organisations (1)
Abstract
How good is the survival of our observed group when compared to the general population survival? This seemingly simple question has been frequently addressed in the literature and often answered with the number of years lost or saved compared to the general poulation. However, the approaches to estimate this measure are typically highly biased and often misleading due to implicit immortal bias. The number of years of life lost due to a specific cause has also been reported in the literature, but the methodology there is even more questionable as the total of years lost by causes may be greater than the overall number of years lost. Recently, a new estimator of years of life lost or gained has been proposed and its variance has been defined. Further work must be done to make this approach useful in practice.
First, the properties of the measure must be clearly understood and the estimator must be redefined to allow the time-varying hazards in the general population and to take into account the fact that the administrative censoring at the end of the study may be informative due to the long follow-up time. The variance estimator must be adapted to take into account the variability of the population part, which is not independent of the observed data. We will consider pseudo-observations as a tool that may prove useful in modelling the number of years of life saved or lost. We will study the properties of such models and adapt them for left truncation and the presence of covariate dependent informative censoring, as these are two of the issues that are common in data with long follow-up time. Lastly, all the newly proposed methodology requires development of user-friendly software. Our project will address all these issues in a comprehensive way and thus ensure that the methodology will be well understood and ready to be used in standard survival analyses.
Significance for science
The goal of the project is to develop the methodology of years lost or saved compared to the general population in a comprehensive way, so it could be well understood and directly used in practice. The project is based on the recently introduced cause-specific estimator – the basic theory has been developed and the estimator has been used in the analysis of patients with bipolar disorder.
By developing this theory further we wish to make this methodology an important new and generally usable approach that can shed new light on survival data. With a comprehensive approach, we intend to deal with all the major issues that arise in practical use of the methodology.
The basic theoretical advance is to define a consistent estimator in case of time-varying population mortality hazard, to deal with informative censoring and to develop a sensible variance estimator. We expect that regression models with years lost/saved as the outcome will prove as another useful application of the pseudo-observations theory. Furthermore, the left-truncated outcome will present the first practical application of the most recent theory of stopped pseudo-observations.
The development of efficient algorithms and user-friendly functions will enable wide usage of the methodology which will contribute to diminishing the gap between the advances in statistical theory and its use in practice. However, the availability of easy-to-use functions only makes sense when accompanied with a comprehensive study of the properties of the measure that enables the user to make valid interpretation of the results.
The new methodology will provide interesting new information when analyzing survival data and will be particularly interesting when studying relative survival of groups of patients whose survival is comparable or even better than that of the general population – such data can not be analyzed with standard relative survival methodology.
The methodology will be particularly useful in presenting the results of scientific studies to the layman audience as it provides a well defined and easy to understand outcome of analyses.
Significance for the country
The goal of the project is to develop the methodology of years lost or saved compared to the general population in a comprehensive way, so it could be well understood and directly used in practice. The project is based on the recently introduced cause-specific estimator – the basic theory has been developed and the estimator has been used in the analysis of patients with bipolar disorder.
By developing this theory further we wish to make this methodology an important new and generally usable approach that can shed new light on survival data. With a comprehensive approach, we intend to deal with all the major issues that arise in practical use of the methodology.
The basic theoretical advance is to define a consistent estimator in case of time-varying population mortality hazard, to deal with informative censoring and to develop a sensible variance estimator. We expect that regression models with years lost/saved as the outcome will prove as another useful application of the pseudo-observations theory. Furthermore, the left-truncated outcome will present the first practical application of the most recent theory of stopped pseudo-observations.
The development of efficient algorithms and user-friendly functions will enable wide usage of the methodology which will contribute to diminishing the gap between the advances in statistical theory and its use in practice. However, the availability of easy-to-use functions only makes sense when accompanied with a comprehensive study of the properties of the measure that enables the user to make valid interpretation of the results.
The new methodology will provide interesting new information when analyzing survival data and will be particularly interesting when studying relative survival of groups of patients whose survival is comparable or even better than that of the general population – such data can not be analyzed with standard relative survival methodology.
The methodology will be particularly useful in presenting the results of scientific studies to the layman audience as it provides a well defined and easy to understand outcome of analyses.