Levels, Trends and Determinants of Technical Efficiency of General Hospitals in Uganda: Data Envelopment Analysis and Tobit Regression Analysis

Abstract

Background: General hospitals provide a wide range of primary and secondary healthcare services. They accounted for 38% of government funding to health facilities, 8.8% of outpatient department visits and 28% of admissions in Uganda in the financial year 2016/17. We assessed the levels, trends and determinants of technical efficiency of general hospitals in Uganda from 2012/13 to 2016/17. Methods: We undertook input-oriented Data Envelopment Analysis to estimate technical efficiency of 78 general hospitals using data abstracted from the Annual Health Sector Performance Reports for 2012/13, 2014/15 and 2016/17. Trends in technical efficiency was analysed using Excel while determinants of technical efficiency were analysed using Tobit Regression Model in STATA 15.1. Results: The Average Constant Returns to Scale, Variable Returns to Scale and Scale Efficiency of general hospitals for 2016/17 were 49% (95% CI, 44% - 54%), 69% (95% CI, 65% - 74%) and 70% (95% CI, 65% - 75%) respectively. There was no statistically significant difference in the efficiency scores of public and private hospitals. Technical efficiency generally increased from 2012/13 to 2014/15, and dropped by 2016/17. Some hospitals were persistently efficient while others were inefficient over this period. Hospital size, geographical location, training status and average length of stay were statistically significant determinants of efficiency at 5% level of significance. Conclusion: The 69% average variable returns to scale technical efficiency indicates that the hospitals could generate the same volume of outputs using 31% (3,439) less staff and 31% (3,539) less beds. Benchmarking performance of the efficient hospitals would help to guide performance improvement in the inefficient ones. There is need to incorporate hospital size, geographical location, training status and average length of stay in the resource allocation formula and adopt annual hospital efficiency assessments.

Approach in Inputs & Outputs Selection of Data Envelopment Analysis (DEA) Efficiency Measurement in Hospital: A Systematic Review

“…Technical Efficiency (TE), θ CCR = Pure Technical Efficiency (PTE), θ BCC x Scale Efficiency The Fig 2 illustrates geometrically the concepts of efficiency measurement in DEA Twenty nine percent (26 of 89, 29.21%) of the included studies measured the (overall) Technical Efficiency (TE) [77–102] and 26.97% (24 of 89) [75,76,103–124] used TE, PTE and SE in their evaluation. While the rest of the studies evaluated their efficiency by the combination of TE and PTE and some not stated clearly the type of efficiency measured (for details see Appendix C).…”

“…While others detailed the model as Radial, Non-Radial and Oriented, Non-Radial and Non-Oriented, and Radial and Non-Radial [23,126]. The majority of the research that were included used Radial DEA models (72 of 89, 80.90%) [46,48–52,54–62,65–77,79,80,82–86,89–94,96–100,102–116,118–124,127–130]. BCC, CCR or combination of both models were the specific model utilized in measuring efficiency via radial change of input and output values.…”

“…In other words, either output expansion or input decrease is the primary evaluation goal. More than half of the studies (55.06%, 49 of 89) [46–48,50–54,58,60,64–69,71–73,75–77,80,91–94,96,98,99,103–107,109–111,113–116,118,121,124,127,128,130,135] applied input-orientated DEA models. The researchers indicated that the choice of input orientation was made to be in line with the fact that the majority of healthcare units aim to minimize inputs given a goal level of output.…”

“…Among the studies, 80 (89.89%) conducted extended analysis in their research [46–56,58–64,66–69,71–77,79–94,96–98,101–116,118–125,128–135]. The type of data applied were almost equally distributed among the studies 51 (57.30%) [47,49,52,58–60,67,70–73,75–77,79,80,83–85,88,91,92,94,96,97,100,101,103,105–109,111,112,114–116,120–122,124,125,127,129–135] used panel data and 38 (42.70%) [46,48,50,51,53–57,61–66,68,69,74,78,81,82,86,87,89,90,93,95,98,99,102,104,110,113,117–119,123,128] used cross sectional data in their research. Forty extended analyses were identified within the included studies.…”

“…Some of the studies used combination of approaches. 63 studies (70.79%) used only literature review as their approach in selecting the input and output [46–48,50–55,59,60,62,64–78,81,82,85,86,89–92,94,96–98,101,103,105,108,110,111,113,115–122,124,125,127,128,130,132–135]. While the rest used literature review in combination with other approaches.…”