Every method for computing the need for Primary Care Physicians in Alabama must start by defining a collection of service areas that partition the state. The assumption is that patients in a service area will seek care from a PCP in that service area. The service areas could possible be
Counties are most commonly used, but are perhaps too coarse. ZIP code areas and census tracts are too fine. All of the first three are artificial, based on boundaries that have nothing to do with healthcare. PCSAs are the most rational and have the proper resolution. The various area layers can be viewed in the interactive map.
After defining service areas, the next step is a computational model for determining the number of PCPs needed for the service area, and then comparing that with the number of PCPs present in the service area.
The most common model for determining need is based on computing Population to PCP ratio or PCP to Population ratio in the service area. The various ratio methods are equivalent. In this presentation, we use the number of PCPs per 10,000 population so that the numbers are smaller and easier to understand. So for example, the need could be based on a desired PCP per 10K ratio of
Another method is to compute the demand for PCP services, based on the demographics of the service area, and the supply of PCP services, based on the number of PCP FTEs in the service area. Both supply and demand can be based on time or visits (or perhaps other metrics).
The following table gives the mean and the standard error of the number of visits per year (VPY) to a primary care physician (PCP) and the time (in minutes) per visit (MPV) for various age groups. The data are from the National Ambulatory Medical Care Survey, published by the Centers for Disease Control.
| Age | Mean VPY | Standard Error VPY | Mean MPV | Standard Error MPV |
|---|
The total average number of PCP visits needed per year for a service area can be computed by multiplying the population in each age group by the mean visits per year for that age group, and then summing over the 13 age groups. This gives the demand for the service area in visits per year. On the other hand, on average, an FTE PCP sees 20 patients per day, 5 days per week, for 48 weeks per year, for a total of 4800 visits year. This is the supply in visits per year per FTE. The FTE need for a service area is computed by dividing the total number of visits required by the population in a year by 4800, the average number of visits an FTE can supply in a year. Finally then the difference between the FTEs present in the PCSA and FTEs needed is either the surplus (if positive) or deficiency (if negative). \[D = \sum_{i=1}^{13} p_i v_i, \quad S = 5 \cdot 48 \cdot 20 = 4800, \quad N = D / S, \quad \Delta = F - N\]
The total average PCP time needed per year for a service area can be computed by multiplying the population in each age group by the mean visits per year and by the mean time per visit for that age group, and then summing over the 13 age groups. This gives the demand for the service area in minutes per year. On the other hand, on average, an FTE PCP works 8 hours per day, 5 days per week, and 48 weeks per year, for a total of 115,200 minutes of patient time per year. This is the supply in minutes per year per FTE. The FTE need for a service area is computed by dividing the total time required by the population in a year by 115,200, the average yearly patient time of an FTE. Once again, the difference between the FTEs present in the PCSA and the FTEs needed ie either the surplus (if positive) or the deficiency (if negative). \[D = \sum_{i=1}^{13} p_i v_i t_i, \quad S = 5 \cdot 48 \cdot 8 \cdot 60 = 115200, \quad N = D / S, \quad \Delta = F - N \]