Using gini-style indices to evaluate the spatial patterns of health practitioners: Theoretical considerations and an application based on Alberta data

https://doi.org/10.1016/0277-9536(94)90189-9Get rights and content

Abstract

The paper analyzes how Gini-style indices are optimally used in the evaluation of economic spatial models designed to predict where health care practitioners are likely to locate under competitive market conditions. At a conceptual level, the analysis establishes that Gini-style indices can be brought to bear on economic models, only if the ordering of geographic areas required to give Gini-coefficient values internal technical coherence also has meaning in terms of the conceptual predictions of the modelling. This, in turn, implies that Gini-indices are most likely to prove useful for fairly aggregated forms of economic analysis, involving relatively few and large geographic divisions. At an applied level, the analysis establishes that one particular geographic distribution of health practitioners is empirically dominant, and that is the distribution which involves the lowest practitioner: population ratio in rural areas, and the highest ratio in large urban areas, with the ratio for small urban areas in between. The empirical evidence also suggests that the spatial practitioner distributions are highly stable for most kinds of health personnel, making it problematic whether these distributions can be changed through normal types of public policy interventions.

References (19)

There are more references available in the full text version of this article.

Cited by (200)

  • Reconstructing mutational lineages in breast cancer by multi-patient-targeted single-cell DNA sequencing

    2023, Cell Genomics
    Citation Excerpt :

    Amplicon sequencing can have a relatively non-linear distribution of read depth, so we examined the coverage read depth distributions of each tumor sample to access the extent of this variance (Figure S2B). Summary Gini index (GI ) (gini()) and coefficient of variation (CV) (SD/mean) were calculated for both the cells and amplicons by taking the average of both metrics for each sample (average of all cells/amplicons per sample) (Figures S1A and S1B, Table S3).54 To obtain a higher resolution, we next calculated the GI of each respective amplicon per sample to better understand the individual contribution of each amplicon to the variance (Figure S2A).

View all citing articles on Scopus
View full text