SSaúde_Lecture11: Resource Allocation in Health
Data Flow in Health Care Organizations
• Health analytics deal with big datasets
which often flow through multiple systems
before they are in a usable format for
health care managers.
• Once the data are in a usable format,
health care managers can perform their
analyses to support various operational
and strategic decisions.
Sofware: MANIPULATING DATA
• Health care managers can import data through various means into data analysis and
visualization tools such as Excel and Tableau.
• Often when data is imported into a data analysis tool, it requires additional clean -
up and manipulation.
• The PivotTable functionality in Excel is a helpful tool for transforming the dataset
into a workable format.
Resource allocation in health with Optimization
Context Questions:
• Which mix of health care services maximizes profit, taking into account the
availability of medical and non-medical resources?
• How should different gamma cameras be used to make exams so as to m inimize
costs, and taking into account the time required to perform a ll the exams and the
time required for maintenance?
• Which is the minimal number of doctor shifts required to ensure that all patients are
observed in an emergency unit, knowing that doctors can only work 40 hours per
week and the maximum of 12h in a row?
• How to approach this?
Optimization through linear programming
Linear programming (LP) models are excellent tools to solve allocation, p urchase-mix,
scheduling problems, etc…
1
A large number of decision variables can be incorporated into a single model to attain an
optimal solution to a problem.
Structure of an LP:
Contains 4 components:
• Objective
o Either maximization or minimization; a mathematical statement of profit,
cost, etc. per unit of output or input
• Decision Variables
o Amount of either inputs or outputs
• Constraints
o Limitations that restrict the available alternatives
• Parameters
o Numerical values (eg: values of available resources)
General structure of the LP is as follows:
2
Typical objective functions:
• Maximize profit
• Minimize costs
• Maximize turnover
• Maximize return on investment
• Minimize distances
• Maximize access
• Minimize net present value
• Maximize utility
• Maximize health gains
Maximization models:
• Maximization models seek either revenue or profit maximization, health gains
maximization, etc.
• To illustrate these concepts in an example, and build a LP model f or the example,
consider the following maximization example:
o VER PPT
Minimization models:
• When the measures in the objective function are the costs, obviously healthcare
managers seek to minimize the costs
• Model set-up follows the same steps, with one exception: in cost minimization
problems, the constraints are generally the ≥ type. Thus, in the graphic solution, the
feasible area is defined from infinity towards origin .
o VER PPT
Integer Linear Programming
In LP one of the assumptions is that decision variables are continuous. Therefore solutions
can yield fractional values such as 4.3 patients or 7.6 nurses.
Such solutions can be, in some contexts impractical, however, when LP is used for
scheduling clinical staff rounding off these values may generate infeasible or less optimal
solutions.
Integer programming is an extension of LP that eliminates the problem by enforcing integer
decision variable outcomes.
VER PPT (exemplo)
