Probabilistic population projections for countries with generalized HIV/AIDS epidemics.

Area of Science:

• Demography
• Epidemiology
• Public Health

Background:

• The United Nations (UN) provides probabilistic population projections up to 2100.
• Mortality projections for countries with generalized HIV/AIDS epidemics typically use the complex Spectrum/Estimation and Projection Package (EPP) model.
• The EPP model is designed for short-term, policy-relevant epidemic projections and is not fully compatible with standard UN projection methods.

Purpose of the Study:

• To propose a simpler, more compatible method for probabilistic population projections in countries affected by HIV/AIDS.
• To integrate life expectancy changes into the standard cohort component projection method used by the UN.
• To provide an alternative to the complex multistate modeling of the Spectrum/EPP model.

Main Methods:

• Projecting changes in life expectancy probabilistically using time series regression.
• Converting projected life expectancy changes into age- and sex-specific mortality rates via model life tables for HIV/AIDS-affected populations.
• Inputting these mortality rates into the standard UN cohort component projection method.

Main Results:

• The proposed method demonstrated good performance in out-of-sample cross-validation.
• Short-run population projections align closely with those generated by the Spectrum/EPP model.
• The new approach is simpler and avoids the need for multistate modeling.

Conclusions:

• A simplified probabilistic approach for mortality projections in HIV/AIDS-affected countries is feasible and effective.
• This method enhances compatibility with existing UN population projection frameworks.
• The proposed method offers a more accessible alternative to complex epidemic modeling for long-term population projections.