jackolney / CareCascade

CareCascade Model Repository
0 stars 0 forks source link

LogicalThinking #30

Closed jackolney closed 9 years ago

jackolney commented 9 years ago

The whole premise of this work was to demonstrate how intervening at a single point in care is silly; as due to upstream / downstream weaknesses in care you will never truly realise the full impact of your intervention.

To test this, we need to bolster care. So, we should sequentially make care perfect and then show the impact of our interventions

In other words, for a specific intervention, I want to make care upstream perfect, see the impact of the intervention, then make care downstream perfect and see the impact of the intervention

Look at the impact of our interventions on:

  1. Excess mortality
  2. DALYs

Let the narrative of the paper guide the analysis

jackolney commented 9 years ago

Making a connection between how much each intervention patches the 'leak' in the cascade that it aims to address and how much it reduces the overall HIV health losses

jackolney commented 9 years ago

PLAN

Build up 'perfect' care and then sequentially weaken care from that point to illustrate the importance of each aspect of care

In the absence of interventions (no HBCT either)


To begin with, to make 'perfect care':

I don't think I need to be altering tx guidelines Additionally, ignore 2010, just start from 2004...

jackolney commented 9 years ago

Everything kinda grinds to a halt if you set the VCT test time to 1 day...

Try only giving people one entrance to care... They don't need to keep reappearing if they get in everytime