The Validity of “ Cheap , Fast , Good : Pick Any Two ” in Evaluating Healthcare Systems
of a factor as the Balanced Providers percentage increased . A different challenge occurred with Balanced Providers at 15 %; as this is not a number that evenly divides into 100 , it may be noted that the total number of Providers does not directly match with the percentage one would expect with 100 consumers . In these cases , this must be accepted as an inevitable rounding error . After obtaining results for 1000 consumers , the model was run again , this time lowering the number of consumers to 800 , 600 , 400 , and 200 . In each case , the model was run until an inflection was reached , for the same number of balanced providers as with 1000 consumers . This was done to determine how big of a role scale played , as it had been observed that varying the number of balanced providers had a significant effect . With this in mind , the range for each the inflection point is shown in Table 1 . In this table , two things become clear . One is that as the percent of Balanced Providers increases , the Inflection Point occurs at a higher ratio of Consumers to Providers , rising from a range of 6.30 — 6.41 with zero Balanced Providers to a range of 10.42 — 10.87 with 25 %
Balanced Providers for 1000 consumers . The second thing that becomes clear is that the inflection point generally trends lower for lower numbers of consumers . There are some exceptions , such as 25 % Balanced Providers and 800 vs 600 Consumers , but these are rare . Note that Table 1 only shows the inflection points ; it does not show the other runs above or below those points that helped narrow down the inflection points . Figures 3 and 4 show the upper and lower values of the inflection points , respectively . In these figures , the x-axis represents the percentage of Balanced Providers while the y-axis represents the ratio of Consumers to Providers .
Having noted that the inflection point increases with the increase in Balanced Providers , what does this actually mean ? Among others , one thing this indicates is that having more , varied options for consumers fills a vacant niche , which makes the market more robust , and benefits all providers . The follow-on question then becomes the following : if a Cheap , Fast , Good provider can exist in a model , why do we say that it does not or cannot exist in
Table 1 . Inflection Points , 200-1000 Consumers
Percent Balanced Providers
Inflection Point , 1000 Consumers
Inflection Point , 800 Consumers
Inflection Point , 600 Consumers
Inflection Point , 400 Consumers
Inflection Point , 200 Consumers
0 6.30 -- 6.41 6.30 -- 6.45 5.88 -- 6.06 5.80 -- 6.06 5.13 -- 5.55 5 7.14 -- 7.30 7.21 -- 7.34 7.23 -- 7.50 6.67 -- 7.02 5.71 -- 6.25 10 8.33 -- 8.70 8.0 -- 8.33 7.79 -- 8.11 8.0 -- 8.51 6.67 -- 7.41 15 9.09 -- 9.43 8.70 -- 8.99 8.82 -- 9.38 8.51 -- 9.30 7.41 -- 8.33 20 9.52 -- 9.80 9.64 -- 10.0 9.38 -- 10.0 8.89 -- 9.30 8.0 -- 9.09 25 10.42 -- 10.87 10.0 -- 10.53 10.0 -- 10.71 10.0 -- 11.111 8.33 -- 10
89