Measuring Productivity and Relative Efficiency of Public Tertiary Traditional Chinese Medicine Hospitals in Hubei,China

Abstract

Improving the productivity and relative efficiency of traditional Chinese medicine (TCM) hospitals is pivotal for hospital managers and policymakers to optimize the utilization of TCM resources in China. This study aimed to measure the productivity and relative efficiency of public tertiary TCM hospitals in Hubei Province. The input and output indicators data were extracted from the Health Commission of Hubei Province (HCHP) from 2019 to 2021. The Bootstrap-Malmquist-DEA model was employed to measure the productivity and relative efficiency of the hospitals. The statistical significance was set at P < .05. The numbers of total diagnostic patients and discharged patients declined by 23.44% and 28.34% from 2019 to 2020, and then increased by 25.76% and 20.44% respectively from 2020 to 2021. The average bias-corrected technical efficiency (TE) scores of the TCM hospitals from 2019 to 2021 were 0.8391, 0.8048, and 0.8559, indicating good efficiency. The average total factor productivity (TFP) in 2020 and 2021 decreased compared to that in 2019, with scores of 0.7479 and 0.8996, respectively. Between 2019 and 2020, the TFP changes among 19 out of 21 (90.48%) TCM hospitals and the technological changes (TC) among 20 out of 21 (95.24%) were less than 1.0000 (P < .05). The TFP changes of 17 out of 21 (80.95%) TCM hospitals and the TC of 20 out of 21 (95.24%) were less than 1.0000 (P < .05) between 2019 and 2021. COVID-19 might have constrained the provision of healthcare services by the public tertiary TCM hospitals in Hubei Province. Priority should be given to the utilization of healthcare resources, performance evaluation, information system strengthening, and internal hospital management to boost technical efficiency. TCM hospitals need to focus further on technology innovation to improve their technological progress.

Keywords

traditional Chinese medicine hospitals Bootstrap-Malmquist-DEA model productivity relative efficiency technical efficiency technological progress

What do we already know about this topic?

Several studies have been conducted using classical DEA models to measure the productivity and relative efficiency of TCM hospitals in China. However, incorporating all levels of TCM hospitals simultaneously, including primary, secondary and tertiary TCM hospitals, can lead to biased results due to heterogeneity. The scores generated from the classical models need to be adjusted using a Bootstrapping approach.

How does your research contribute to the field?

The COVID-19 pandemic might have constrained the provision of healthcare services by the public tertiary TCM hospitals in Hubei Province of China over the study period. It might also affect their productivity and relative efficiency. In terms of the total factor productivity changes for the TCM hospitals, the technological changes were the main driver, indicating an imperative need for innovation in technology.

What are your research’s implications toward theory, practice, or policy?

Priority should be given to the utilization of healthcare resources, performance evaluation, information system strengthening, and internal hospital management to boost technical efficiency. TCM hospitals need to focus on further innovating technology to improve their technological progress.

Introduction

The Healthy China 2030 Planning Outline, approved by China’s Party Committee and the State Council, has proposed 2 important goals: providing efficient and high-quality medical services and fully leveraging the potential of Traditional Chinese Medicine (TCM).¹ TCM development has been significantly improved as a crucial aspect of Chinese healthcare since the implementation of the new medical reform, due to numerous TCM-focused policies released by the Chinese government. Consequently, TCM services have become more accessible and affordable, thereby alleviating the medical burden.² To be noteworthy, TCM has made valuable contributions to the prevention and treatment of COVID-19, a challenging global health issue. Gajewski et al conducted a literature review and reported that over 85% of COVID-19 patients in China used TCM, and the most common herb (Glycyrrhiza glabra), when combined with standard therapies, significantly reduced hospitalization rates and COVID-19 symptom occurrence.³

TCM hospitals paly a dominant role in providing TCM services in China.⁴ However, these hospitals face various challenges, such as lower pricing of TCM outpatient services compared to Western medical services,⁵ significant loss of human resources, and an imbalanced structure of health professionals.⁶ Moreover, a majority of TCM programs are not covered by public health insurance schemes, which further discourages patients from choosing TCM as their primary option.⁷ Hence, it is critical for TCM hospitals to provide high-quality TCM services. One key aspect of improving TCM service quality is to promote the operational efficiency of TCM hospitals.⁸ However, TCM resources are often limited. Therefore, it is pivotal for hospital managers and policymakers to prioritize enhancing the productivity and efficiency of TCM hospitals in order to optimize the utilization of TCM resources in China.

Data Envelopment Analysis (DEA) has been widely applied to calculate the productivity and relative efficiency of healthcare institutions due to its ability to handle multiple inputs and outputs.⁹ The Malmquist Productivity Index (MPI) model is also employed to assess the dynamic changes in productivity and relative efficiency of healthcare institutions or health systems.^10,11 Several studies have used the DEA/MPI models to measure the productivity and relative efficiency of TCM hospitals in China.^12-15 However, one limitation is that they incorporated all levels of TCM hospitals simultaneously, including primary, secondary and tertiary TCM hospitals, which resulted in biased results due to the heterogeneity across these different levels. Based on the Hospital Classification Management Measures in China, tertiary TCM hospitals represent the highest level and vary in capacity, structure of health professionals, and hospital settings.¹⁴ Besides, the scores generated by classical DEA/MPI models were not adjusted, further contributing to bias. To address these limitations, Simar and Wilson suggest combining the classical DEA/MPI models with a Bootstrapping strategy to account for environmental and random factors.^16,17 Therefore, this study aimed to measure the productivity and relative efficiency of public tertiary TCM hospitals in Hubei Province of China, using the Bootstrap-Malmquist-DEA model from both horizontal and longitudinal perspectives. The findings can provide empirical evidence for hospital managers and policymakers to improve the management and quality of TCM medical services.

Methods

Study Design

Figure 1 demonstrates the design to measure the productivity and relative efficiency scores of public tertiary TCM hospitals in Hubei Province. A horizontal analysis and a longitudinal analysis were undertaken using a Bootstrap-DEA model and a Bootstrap-Malmquist-DEA model, respectively. Technical efficiency scores were obtained from the Bootstrap-DEA model for the cross-sectional comparisons, while total factor productivity changes (TFPC) were generated from the Bootstrap-Malmquist-DEA model for longitudinal comparisons.

Figure 1.

Flow chart of measuring the productivity and relative efficiency of public tertiary TCM hospitals.

Input and Output Indicators

A systematic review on relative efficiency and productivity research of Chinese hospitals using DEA models conducted by Dong et al,⁹ suggested that there is often a misconception regarding technical efficiency and allocative efficiency due to the inappropriate combination of monetary and volume indicators. This review, along with a previous study published by our research team,¹⁸ also highlighted the challenge of accessing price information compared to volume information in hospitals. Therefore, in this study, we selected input and output indicators based on their frequency of use in current studies and data feasibility. The input indicators chosen were the number of health professionals and the number of beds, while the output indicators included the number of total diagnostic patients, and the number of discharged patients.

Bootstrap-Malmquist-DEA Model

In a horizontal analysis, the Bootstrap-DEA model was used to estimate the relative technical efficiency (TE) of public tertiary TCM hospital in Hubei Province. In a longitudinal analysis, the Bootstrap-Malmquist-DEA model was employed to measure the dynamic changes in productivity and relative efficiency. Specifically, the TFPC and its decomposition index, including technical efficiency changes (TEC) and technological changes (TC), assuming that all decision-making units (DMUs), that is, TCM hospitals, are operating on the frontier.¹⁹ Additionally, TC can be further decomposed into pure technical efficiency changes (PTEC) and scale efficiency changes (SEC). The basic formula is as follows: TFPC=TEC×TC= (PTEC×SEC) ×TC. Generally, if the productivity and relative efficiency scores are greater than 1.0000, it indicates an increase; if the scores are equal to 1.0000, it implies stability; and if the scores are less than 1.0000, it suggests a decrease.

Considering the productivity and relative efficiency of all DMUs, their operations are actually affected by environmental and random factors, the Bootstrap method was introduced as an approach to further address this issue based on classical DEA models.^9,16,17,19 Its basic idea is to simulate the data-generating process (DGP) through repeated sample selection, thereby generating bias-corrected productivity and relative efficiency scores with 95% confidence intervals (CIs). In result, the obtained scores are much closer to the actual productivity and relative efficiency of TCM hospitals. Furthermore, the Bootstrap-Malmquist-DEA model is able to infer whether changes in productivity and relative efficiency are statistically significant.

Data Sources, Extraction and Processing

The data for the input and output indicators were extracted from the Health Commission of Hubei Province (HCHP) from 2019 to 2021. The respective number of the public tertiary TCM hospitals was 21, 23, and 24 from 2019 to 2021. Besides, in this study, the public tertiary TCM hospitals were labeled with the sequence T1, T2, and T3 and so on for each year. According to the classification of bias-corrected TE scores, 5 categories were created, including excellent efficiency [0.9000, 1.0000], good efficiency [0.8000, 0.9000], average efficiency [0.7000, 0.8000], poor efficiency [0.6000, 0.7000], and failing efficiency [0.000, 0.6000].^18,20 Besides, the ranking orders were determined based on the bias-corrected TE scores.

Statistical Analysis

Descriptive analysis of the input and output indicators, including mean and standard deviation (SD), was conducted using SPSS 23.0 version statistical. The productivity and relative efficiency of the public tertiary TCM hospitals were measured using R 3.2.1 version statistical software, specifically employing FEAR package with 95% CIs.²¹ The geometric mean was used to represent the average scores of productivity and relative efficiency. A statistical significance was determined if the 95% CIs did not contain the value of 1.0000. The number of repeated sample selection was set at 2000. The statistical significance was set at P < .05.

Results

Characteristics of the Input and Output Indicators for the Public Tertiary TCM Hospitals From 2019 to 2021

Table 1 shows that regarding the input indicators, the numbers of health professionals and beds varied slightly from 2019 to 2021. The highest numbers of health professionals and beds in 2021 were 856 and 790, respectively. As to the output indicators, the numbers of total diagnostic patients and discharged patients experienced a decline of 23.44% and 28.34% from 2019 to 2020. However, the numbers of total diagnostic patients and discharged patients increased by 25.76% and 20.44% from 2020 to 2021.

Table 1.

Summary of the Input and Output Indicators for the Public Tertiary TCM Hospitals From 2019 to 2021.

Year	Number of TCM hospitals	Input indicators		Output indicators
		Mean ±SD
		Number of health professionals	Number of beds	Number of total diagnostic patients	Number of discharged patients
2019	21	824 ± 456	786 ± 353	480 464 ± 525 915	26 260 ± 14 068
2020	23	820 ± 428	790 ± 403	367 799 ± 330 648	18 817 ± 9660
2021	24	856 ± 503	790 ± 413	462 532 ± 508 121	22 663 ± 13 712

Technical Efficiency Scores and Ranking After Bias-Correction Among the Public Tertiary TCM Hospitals From 2019 to 2021

As displayed in Table 2, the average bias-corrected TE scores of the public tertiary TCM hospitals from 2019 to 2021 were 0.8391, 0.8048, and 0.8559, belonging to good efficiency. The highest bias-corrected TE scores were 0.9590, 0.9691, and 0.9794, while the lowest bias-corrected TE scores were 0.5519, 0.4682, and 0.6224 for each respective year. Specifically, in 2019 and 2020, only 2 TCM hospitals had failing efficiency scores, accounting for 9.52% and 8.70% of the total, respectively. The numbers of TCM hospitals with excellent and good efficiency scores from 2019 to 2021 were 17 (80.95%), 18 (78.26%) and 18 (75.00%), respectively.

Table 2.

Technical Efficiency Scores and Ranking After Bias-Correction Among the Public Tertiary TCM Hospitals From 2019 to 2021.

Year	No	Before bias-correction	After bias-correction	Bias	Lower bounds	Upper bounds	Ranking orders
2019	T1	1.0000	0.8970	−0.1148	0.7043	0.9986	11
	T2	1.0000	0.8949	−0.1175	0.6985	0.9991	12
	T3	0.9566	0.9242	−0.0366	0.8600	0.9556	7
	T4	0.6106	0.5867	−0.0666	0.5389	0.6098	20
	T5	0.5688	0.5519	−0.0539	0.5087	0.5682	21
	T6	0.8589	0.8303	−0.0402	0.7684	0.8581	16
	T7	1.0000	0.9577	−0.0441	0.8359	0.9989	2
	T8	0.6933	0.6704	−0.0492	0.6169	0.6922	19
	T9	1.0000	0.9590	−0.0427	0.8887	0.9991	1
	T10	0.9401	0.9080	−0.0376	0.8508	0.9390	9
	T11	1.0000	0.9353	−0.0692	0.8206	0.9990	4
	T12	0.8809	0.8563	−0.0327	0.8066	0.8799	14
	T13	0.8443	0.8258	−0.0264	0.7896	0.8433	17
	T14	1.0000	0.9265	−0.0793	0.8366	0.9987	6
	T15	0.9004	0.8808	−0.0248	0.8302	0.8997	13
	T16	1.0000	0.9374	−0.0668	0.8622	0.9990	3
	T17	0.8602	0.8370	−0.0323	0.7879	0.8594	15
	T18	1.0000	0.9317	−0.0733	0.8372	0.9988	5
	T19	0.9253	0.9031	−0.0265	0.8549	0.9245	10
	T20	0.9366	0.9147	−0.0256	0.8571	0.9357	8
	T21	0.7207	0.6914	−0.0588	0.6309	0.7198	18
2020	T1	1.0000	0.8823	−0.1334	0.6918	0.9987	11
	T2	1.0000	0.8808	−0.1354	0.6640	0.9987	12
	T3	0.8698	0.8314	−0.0532	0.7618	0.8690	13
	T4	0.4932	0.4682	−0.1082	0.4152	0.4926	23
	T5	0.5219	0.5020	−0.0761	0.4608	0.5212	22
	T6	0.8297	0.8006	−0.0438	0.7433	0.8286	18
	T7	1.0000	0.9245	−0.0817	0.7745	0.9987	3
	T8	0.6287	0.6054	−0.0612	0.5559	0.6280	21
	T9	0.6678	0.6390	−0.0675	0.5944	0.6670	20
	T10	1.0000	0.9062	−0.1035	0.7516	0.9985	6
	T11	1.0000	0.9004	−0.1107	0.7590	0.9987	8
	T12	0.8656	0.8240	−0.0582	0.7499	0.8642	15
	T13	1.0000	0.9298	−0.0756	0.8073	0.9987	2
	T14	0.8452	0.8285	−0.0239	0.7875	0.8444	14
	T15	0.8274	0.8066	−0.0313	0.7664	0.8265	16
	T16	0.9614	0.9213	−0.0453	0.8520	0.9602	4
	T17	1.0000	0.9128	−0.0955	0.8132	0.9988	5
	T18	0.9950	0.9691	−0.0269	0.8937	0.9944	1
	T19	0.8244	0.8029	−0.0325	0.7564	0.8234	17
	T20	0.9303	0.8930	−0.0449	0.8219	0.9289	10
	T21	0.9326	0.9037	−0.0343	0.8444	0.9317	7
	T22	0.9133	0.8949	−0.0225	0.8521	0.9124	9
	T23	0.8220	0.7884	−0.0519	0.7094	0.8210	19
2021	T1	1.0000	0.9084	−0.1008	0.7312	0.9991	13
	T2	1.0000	0.9077	−0.1017	0.7311	0.9989	14
	T3	0.9300	0.9146	−0.0181	0.8825	0.9292	12
	T4	0.6312	0.6224	−0.0224	0.5986	0.6307	24
	T5	0.6838	0.6686	−0.0333	0.6363	0.6831	22
	T6	0.9143	0.8935	−0.0255	0.8327	0.9134	15
	T7	1.0000	0.9640	−0.0374	0.8366	0.9991	3
	T8	0.7492	0.7318	−0.0316	0.6929	0.7485	21
	T9	0.6707	0.6521	−0.0425	0.6125	0.6700	23
	T10	1.0000	0.9399	−0.0639	0.8081	0.9993	8
	T11	0.9981	0.9738	−0.0250	0.9090	0.9972	2
	T12	0.9644	0.9372	−0.0301	0.8759	0.9637	9
	T13	1.0000	0.9538	−0.0484	0.8360	0.9990	4
	T14	0.8715	0.8469	−0.0333	0.8003	0.8706	16
	T15	0.8504	0.8363	−0.0198	0.7983	0.8497	17
	T16	1.0000	0.9426	−0.0608	0.8586	0.9990	5
	T17	0.9749	0.9402	−0.0378	0.8518	0.9740	7
	T18	1.0000	0.9426	−0.0609	0.8732	0.9991	6
	T19	0.8347	0.8122	−0.0331	0.7557	0.8339	18
	T20	0.7599	0.7396	−0.0360	0.6782	0.7592	20
	T21	1.0000	0.9336	−0.0711	0.8351	0.9993	10
	T22	0.9955	0.9794	−0.0165	0.9347	0.9945	1
	T23	1.0000	0.9329	−0.0719	0.8449	0.9991	11
	T24	0.7680	0.7457	−0.0390	0.6883	0.7673	19

Trends of Changes on the TFP and Its Decomposition Scores Among the Public Tertiary TCM Hospitals From 2019 to 2021

As presented in Table 3, compared with the average TFP in 2019, both in 2020 and 2021 decreased, with scores of 0.7479 and 0.8996, respectively. However, the average TFP in 2021 increased slightly compared to that in 2020, reaching 1.1856. Particularly, between 2019 and 2020, the TFP changes among 19 out of 21 (90.48%) TCM hospitals and the TC among 20 out of 21 (95.24%) were less than 1.0000 with a statistical significance (P < .05). Similarly, the TFP changes of 17 out of 21 (80.95%) TCM hospitals and the TC of 20 out of 21 (95.24%) were also statistically significant and less than 1.0000 (P < .05) between 2019 and 2021. In contrast, between 2020 and 2021, the TFP changes of 20 out of 23 (86.96%) TCM hospitals were statistically significant and more than 1.0000 (P < .05).

Table 3.

Changes of the TFP and Its Decomposition Scores of the Public Tertiary TCM Hospitals From 2019 to 2021.

Year	No	TFPC			TEC			TC			PTEPC			SEC
Year	No	Scores	Lower bounds	Upper bounds	Scores	Lower bounds	Upper bounds	Scores	Lower bounds	Upper bounds	Scores	Lower bounds	Upper bounds	Scores	Lower bounds	Upper bounds
2019-2020	T1	0.6800*	0.6683	0.7008	1.0000	0.8010	1.1508	0.6800*	0.5651	0.7896	1.0000	0.7913	1.1377	1.0000	1.0000	1.0777
	T2	0.6022*	0.5090	0.6172	0.8498*	0.6662	0.9042	0.7087*	0.6392	0.7972	1.0000	0.7390	1.1228	0.8498	0.6730	1.0254
	T3	0.7148*	0.7031	0.7265	0.8614*	0.6894	0.9336	0.8299*	0.7563	0.9639	0.9094*	0.7346	0.9831	0.9472	0.8072	1.0661
	T4	0.6420*	0.6167	0.6497	0.8308*	0.7037	0.9287	0.7728*	0.6696	0.8634	0.8161*	0.6547	0.9061	1.0180	0.9011	1.1200
	T5	0.7418*	0.7381	0.7440	0.9842*	0.8780	1.0767	0.7537*	0.6750	0.8278	0.9175	0.7915	1.0076	1.0727	0.9331	1.2161
	T6	0.6570*	0.5750	0.6751	0.9247*	0.7381	0.9976	0.7105*	0.6433	0.7832	0.9660	0.8506	1.0547	0.9572	0.7842	1.0500
	T7	0.6972*	0.6576	0.7062	0.9458	0.7250	1.0025	0.7371*	0.6852	0.8567	1.0000	0.7792	1.0799	0.9458	0.8364	1.0753
	T8	0.6683*	0.6235	0.6817	0.8984*	0.7814	0.9727	0.7440*	0.6699	0.8144	0.9068	0.7863	1.0212	0.9907	0.8239	1.1025
	T9	0.8133*	0.7449	0.8825	1.0042	0.8441	1.0748	0.8099*	0.7449	0.9271	1.0000	0.3013	1.0690	1.0042	0.9261	1.3902
	T10	0.8540*	0.7689	0.9280	1.0938	0.9050	1.1994	0.7808*	0.7268	0.8642	1.0637	0.6341	1.1408	1.0283	1.0024	1.2636
	T11	0.7080*	0.6254	0.7622	0.9342	0.7522	1.0379	0.7578*	0.6945	0.8326	0.8656	0.7524	1.0287	1.0793	0.8140	1.1772
	T12	0.6534*	0.5976	0.6845	0.8490*	0.6695	0.9255	0.7696*	0.7027	0.8730	0.9595	0.8706	1.0494	0.8849*	0.6807	0.9736
	T13	0.7967*	0.7581	0.8230	1.0319	0.8370	1.1272	0.7721*	0.6997	0.8774	0.9801	0.8758	1.0539	1.0528	0.8518	1.1754
	T14	0.6867*	0.6757	0.7086	0.9113	0.7983	1.0096	0.7536*	0.6801	0.8332	0.9614	0.8620	1.1249	0.9479*	0.7970	0.9911
	T15	1.0174	0.8785	1.0836	1.2564	0.9023	1.4071	0.8097*	0.7294	0.9237	1.1106	0.7477	1.1681	1.1313	1.0447	1.3058
	T16	0.9901	0.9212	1.0090	1.3647	1.1708	1.4959	0.7255*	0.6475	0.7930	0.9950	0.9114	1.1592	1.3715	1.0227	1.5161
	T17	0.8578*	0.8521	0.8700	0.9786	0.7263	1.0744	0.8766	0.7851	1.0548	0.9584	0.8598	1.0348	1.0211	0.7581	1.1154
	T18	0.6841*	0.6357	0.7347	0.9052	0.7799	1.0328	0.7558*	0.6599	0.8309	0.9535	0.8477	1.1168	0.9493	0.7977	1.0564
	T19	0.8344*	0.8011	0.8385	1.1255	0.9811	1.2747	0.7413*	0.6208	0.8198	1.0079	0.8757	1.0907	1.1167	1.0070	1.2771
	T20	0.7031*	0.6908	0.7228	0.9541	0.7881	1.0198	0.7369*	0.6860	0.8420	0.9752	0.8778	1.0600	0.9784	0.7889	1.0772
	T21	0.8617*	0.8517	0.8685	1.1553	0.9944	1.3177	0.7458*	0.6250	0.8324	1.1406	0.8190	1.2965	1.0129	0.9820	1.2239
2020-2021	T1	1.4071*	1.3770	1.4130	1.0000	0.8331	1.2013	1.4071*	1.0406	1.5932	1.0000	0.8356	1.2231	1.0000	0.8889	1.0000
	T2	1.4597*	1.4051	1.5951	1.1613*	1.0733	1.3754	1.2569*	1.0263	1.3403	1.0000	0.7958	1.2622	1.1613	0.8060	1.4083
	T3	1.1001*	1.0950	1.1188	1.0450	0.9592	1.2133	1.0527	0.8449	1.1358	1.0692*	1.0183	1.2305	0.9774	0.8366	1.0692
	T4	1.1995*	1.1898	1.2110	1.0506	0.8753	1.1739	1.1417	0.9905	1.3059	1.2797*	1.2091	1.5420	0.8210*	0.6228	0.8402
	T5	1.3640*	1.2319	1.4205	1.2128*	1.0963	1.3311	1.1247*	1.0098	1.1831	1.3102*	1.2113	1.4835	0.9257	0.7688	1.0096
	T6	1.3768*	1.3358	1.4103	1.0911*	1.0032	1.2576	1.2619*	1.0220	1.3527	1.1049*	1.0277	1.2149	0.9874	0.8820	1.1323
	T7	1.2877*	1.2737	1.3584	1.0467	0.9841	1.2470	1.2303	0.9860	1.3012	1.0000	0.9196	1.2001	1.0467	0.8786	1.1743
	T8	1.3777*	1.3645	1.4211	1.2195*	1.1506	1.3451	1.1297*	1.0170	1.1947	1.1917*	1.0778	1.3465	1.0233	0.8834	1.1457
	T9	1.1149*	1.1076	1.1227	0.9954	0.9211	1.0927	1.1201	0.9997	1.1947	1.0043	0.9328	1.1135	0.9911	0.9357	1.0209
	T10	1.0778*	1.0717	1.1107	0.9961	0.9359	1.1161	1.0820	0.9464	1.1490	1.0000	0.8773	1.2215	0.9961	0.7536	1.1085
	T11	0.8093*	0.7364	0.8954	0.7015*	0.6176	0.8036	1.1537*	1.0318	1.2414	1.0000	0.9065	1.2499	0.7015*	0.5049	0.7625
	T12	1.3243*	1.3043	1.3347	1.1287*	1.0595	1.2769	1.1733	0.9872	1.2409	1.1142*	1.0308	1.2472	1.0130	0.9726	1.0620
	T13	1.0847*	1.0802	1.0942	0.9900	0.9201	1.0906	1.0957	0.9763	1.1680	1.0000	0.8989	1.1738	0.9900	0.8176	1.0863
	T14	1.4376*	1.4285	1.4698	1.2253*	1.1509	1.4011	1.1733	0.9883	1.2404	1.0774	0.9840	1.1635	1.1373*	1.0483	1.3142
	T15	1.1934*	1.1534	1.2090	1.0185	0.9568	1.1447	1.1717	0.9898	1.2374	1.0480	0.9592	1.1440	0.9719	0.8770	1.1090
	T16	1.2248*	1.1956	1.2319	1.1025	0.9900	1.2268	1.1110	0.9577	1.2029	1.0402	0.8709	1.1447	1.0599	0.9936	1.1724
	T17	0.9741*	0.9675	0.9958	0.9597	0.8720	1.1551	1.0150	0.7715	1.0997	1.0000	0.9291	1.2128	0.9597	0.8746	1.0128
	T18	1.0516*	0.9801	1.0986	0.9127*	0.8256	0.9989	1.1522*	1.0368	1.2329	1.0050	0.8199	1.0939	0.9081	0.7986	1.0726
	T19	1.1501*	1.1245	1.1819	1.1117*	1.0068	1.3341	1.0345	0.8087	1.1119	1.0644	0.9665	1.1711	1.0444	0.9296	1.2539
	T20	0.8867*	0.8549	0.9459	0.7532*	0.6248	0.8413	1.1773*	1.0661	1.3399	0.8168*	0.7369	0.9276	0.9221*	0.7447	0.9657
	T21	1.2863*	1.2239	1.3158	1.0962	0.8907	1.2008	1.1733*	1.0640	1.3190	1.0722	0.8522	1.1663	1.0224	0.9361	1.0811
	T22	1.3878*	1.3382	1.4048	1.1219*	1.0475	1.2909	1.2370	0.9851	1.3106	1.0949*	1.0048	1.1802	1.0247	0.9346	1.1922
	T23	1.0150*	1.0137	1.0174	0.9379	0.8741	1.0451	1.0822	0.9435	1.1513	0.9343	0.8536	1.0681	1.0039	0.8880	1.0646
2019-2021	T1	0.9601*	0.9476	0.9830	1.0000	0.7603	1.1671	0.9601	0.7709	1.1364	1.0000	0.7603	1.1784	1.0000	0.9811	1.0000
	T2	0.8959*	0.8744	0.8986	0.9824	0.8858	1.0806	0.9120*	0.8038	0.9823	1.0000	0.7258	1.1843	0.9824	0.7558	1.1836
	T3	0.7851*	0.7709	0.8063	0.8993*	0.7813	0.9762	0.8730*	0.7987	0.9666	0.9722	0.9057	1.0885	0.9249*	0.7888	0.9720
	T4	0.7723*	0.7468	0.7805	0.8728*	0.7558	0.9635	0.8848*	0.7734	0.9823	1.0337	0.9727	1.1838	0.8444*	0.6680	0.8874
	T5	0.9895	0.8775	1.0290	1.1823*	1.0560	1.3095	0.8370*	0.7436	0.8867	1.2021*	1.1116	1.3300	0.9835	0.8480	1.0936
	T6	0.8938*	0.8134	0.9292	1.0042	0.8990	1.0887	0.8901*	0.7920	0.9560	1.0674	0.9685	1.1687	0.9408	0.8223	1.0353
	T7	0.8935*	0.8934	0.8976	0.9900	0.9150	1.0852	0.9025*	0.8086	0.9667	1.0000	0.9026	1.1153	0.9900	0.8718	1.0889
	T8	0.9052*	0.8645	0.9246	1.0859	0.9988	1.1937	0.8336*	0.7424	0.8908	1.0806	0.9924	1.1959	1.0049	0.8876	1.1101
	T9	0.8476*	0.7754	0.9121	1.0003	0.9042	1.0989	0.8474*	0.7606	0.9174	1.0000	0.7636	1.0968	1.0003	0.9060	1.1815
	T10	0.9167*	0.8685	0.9507	1.0375	0.9503	1.1310	0.8835*	0.8015	0.9476	1.0259	0.9325	1.1170	1.0114	0.9630	1.0548
	T11	0.9001*	0.8650	0.9255	1.1005*	1.0014	1.2463	0.8179*	0.6970	0.8754	1.0000	0.8918	1.1757	1.1005	0.9111	1.2237
	T12	0.9157*	0.8682	0.9492	1.0403	0.9495	1.1344	0.8802*	0.7991	0.9471	1.0595	0.9837	1.1560	0.9819	0.8719	1.0792
	T13	0.9288*	0.8710	0.9640	1.0510	0.9594	1.1456	0.8838*	0.8021	0.9474	1.0758*	1.0008	1.1651	0.9770	0.8604	1.0777
	T14	0.8548*	0.8318	0.8848	1.0000	0.8923	1.1096	0.8548*	0.7706	0.9275	1.0000	0.8876	1.1371	1.0000	0.8790	1.0407
	T15	1.0619	0.9816	1.1745	1.2058*	1.0367	1.3620	0.8806*	0.8052	0.9677	1.1106	0.9115	1.2003	1.0858	0.9525	1.2321
	T16	1.0445	0.9373	1.0999	1.2360*	1.0674	1.3808	0.8450*	0.7499	0.9017	1.0000	0.8834	1.1191	1.2360*	1.0013	1.4306
	T17	0.9628*	0.9499	0.9813	1.0879	0.9416	1.1758	0.8850*	0.8099	0.9881	1.0393	0.9386	1.1439	1.0468	0.8700	1.1247
	T18	0.5767*	0.5491	0.5941	0.6817*	0.5809	0.7810	0.8459*	0.7148	0.9357	0.7599*	0.7042	0.9040	0.8972*	0.7203	0.9174
	T19	1.0896*	1.0416	1.1011	1.2339*	1.0090	1.3961	0.8831*	0.7499	0.9992	1.0807	0.8300	1.1606	1.1417*	1.0316	1.2915
	T20	0.9702	0.9381	1.0004	1.0705	0.9840	1.1643	0.9063*	0.8106	0.9755	1.0677	0.9570	1.1455	1.0026	0.9081	1.1322
	T21	0.8786*	0.8736	0.8878	1.0836	0.9862	1.2398	0.8108*	0.6780	0.8769	1.0656	0.9518	1.2125	1.0169	0.9936	1.0946

Means the changes of the TFP and it composition index were statistically significant at P < .05.

Discussion

Potential Impact of the COVID-19 Pandemic on the Inputs and Outputs of the Public Tertiary TCM Hospitals

Our study found that the inputs of the TCM hospitals had been relatively stable, while the numbers of both total diganostic patients and discharged patients from the public tertiary TCM hospitals in Hubei Province declined by 23.44% and 28.34% from 2019 to 2020, and then increased by 25.76% and 20.44% from 2020 to 2021. These findings indicate that the COVID-19 pandemic might have contrained on healthcare services provided by the TCM hospitals. The COVID-19 epidemic first broke out in Wuhan, the capital city of Hubei Province, China, in December 2019.^22,23 In March 2020, the World Health Organization (WHO) declared it a pandemic.²⁴ During the early stage and the peak of the COVID-19 pandemic in 2020, many hospitals, including TCM hospitals, were designated to exclusively treat patients infected with COVID-19 due to the high contagion and rapid spread of the virus.²⁵ Moreover, the availability of hospital beds was limited.²⁶ In result, many patients, especially those with chronic diseases, were unable to access healthcare from hospitals as they had before the COVID-19 pandemic. Instead, patients turned to internet hospitals for healthcare services.²⁷ Hence, it led to the decreasing numbers of total diagnostic patients and discharged patients from 2019 to 2020. However, it is encouraging to note that these numbers showed a rise from 2020 to 2021, although they remained lower than those in 2019 in our study. These evidences indicate that the public tertiary TCM hospitals were gradually recovering from the impact of the COVID-19 pandemic by implementing regular and stringent prevention and control measures to prevent the occurrence of COVID-19 infections.

Implications of the Overall Technical Efficiency for the Public Tertiary TCM Hospitals

Our study found that the public tertiary TCM hospitals in Hubei Province demonstrated good TE from 2019 to 2021, as observed from the horizontal perspective. Moreover, only 2 TCM hospitals exhibited failing efficiency in both 2019 and 2020, while the majority of TCM hospitals achieved excellent and good efficiency scores throughout the study period. These findings suggest that the overall performance of the public tertiary TCM hospitals in Hubei Province was commendable from 2019 to 2021. According to a longitudinal study among county-level TCM hospitals in Hubei Province from 2001 to 2017, the average TE score was 0.686, which was much lower than the scores (0.8391, 0.8048, and 0.8559) in our study.¹⁴ Moreover, compared with another productivity and efficiency analysis study among municipal-level TCM hospitals in Gansu Province, the TE of tertiary TCM hospitals was reported to be 0.868, 0.926, and 0.935 from 2017 to 2019, slightly higher than those in our study.²⁸ One possible reason for this disparity could be the difference in the classification of TCM hospitals. Our study solely focused on tertiary TCM hospitals, while the county-level/municipal-level TCM hospitals encompassed primary, secondary, and tertiary hospitals. Different levels of TCM hospitals have varying capacities, with the tertiary hospitals representing highest level. Another contributing factor may be associated with the application of the DEA model. Our study used the Bootstrap-DEA model to estimate the adjusted TE of TCM hospitals, which is considered more accurate than the calculation derived from the classical DEA model used in the county-level/municipal-level TCM hospitals. Despite the overall good efficiency observed in public tertiary TCM hospitals, a small number of hospitals still exhibited failing or poor efficiency in our study. This indicates the need for immediate measures and actions to improve their technical efficiency. Utilization of healthcare resources, performance evaluation, strengthening information systems and internal hospital management can be prioritized to further to enhance technical efficiency.²⁰ It is worth noting that external factors related to regional healthcare resource disparities in Hubei Province may also contribute to the varying efficiency levels observed. Addressing these external factors and implementing effective internal management strategies will be crucial for improving the overall performance of TCM hospitals.

Technological Progress as the Main Driver of the Total Factor Productivity Changes for the Public Tertiary TCM Hospitals

Our study also revealed a decrease in the average TFP of public tertiary TCM hospitals in Hubei Province in 2020 and 2021 compared to 2019, with a slight improvement in TFP observed from 2020 to 2021. This further supports the point that the COVID-19 pandemic might affect the operation of the TCM hospitals in Hubei Province. Previous reports have highlighted the loss of healthcare services in many Chinese hospitals during the late January to March 2020 period due to the COVID-19 pandemic.²⁹ Following the lifting of the lockdown in April, 2020, people in Wuhan and Hubei gradually returned to normal life, and hospitals had to recover regular healthcare services.²⁵ Besides, in our study, the technological progress was found to be the main driver causing the changes of the TFP among the public tertiary TCM hospitals. High medical technology is one prominent component of high-quality healthcare services. It is important to note that TCM hospitals cannot be directly compared to general hospitals, which primarily concentrate on Western medical services in China.⁵ Hence, the TCM hospitals especially need to prioritize the innovation of technology, thereby improving their technological progress. Moreover, TCM has demonstrated effectiveness in the prevention and treatment of patients with COVID-19.³⁰ As TCM is increasingly accepted worldwide, the experience in China may be referential to other countries that possess their own complementary or alternative medicine practices.³¹

Other Implications for Policy and Practice

High productivity and relative efficiency of public tertiary TCM hospitals are strongly linked to the optimal use of healthcare resources.³² Moreover, the optimal use of healthcare resources, in turn, is dependent on a well-structured health professionals. However, the loss of human resources is still a challenge for TCM hospitals. Hence, implementing an advanced performance and salary system can be able to improve the motivation of the TCM health professionals and reduce job turnover. It is also imperative for TCM hospitals to actively participate in the hierarchical diagnosis and treatment system to optimize the utilization of healthcare resources.

Limitations

Our study has some limitations. First, our study only analyzed the productivity and relative efficiency of the public tertiary TCM hospitals in Hubei Province. It did not explain the relationship between the quality and the productivity and relative efficiency, as well as the relationship between the cost and the productivity and relative efficiency. Further study needs to be conducted to explore these relationships. Second, we did not investigate the impact of environmental factors, such as market competition, location, economic development levels, etc. on the productivity and relative efficiency of the hospitals. Future research is necessary to be undertaken using a 3-stage or 4-stage Bootstrap-DEA model. Besides, as WHO declares that the COVID-19 pandemic was no longer a global health emergency in May, 2023, a follow-up study is needed to again specifically measure how COVID-19 continues to affect the productivity and relative efficiency of the TCM hospitals.

Conclusions

The numbers of total diagnostic patients and discharged patients from the public tertiary TCM hospitals in Hubei Province were declined from 2019 to 2020, and then increased from 2020 to 2021. Besides, the average TFP in 2020 and 2021 was decreased compared with that in 2019, and it was slightly higher in 2021 than that in 2020. Our findings indicate that the COVID-19 pandemic might have an impact on patients’ healthcare-seeking behaviors and the operation of the public tertiary TCM hospitals in Hubei Province. A small number of the public tertiary TCM hospitals demonstrated poor or failing technical efficiency. It is suggested that the priority should be given to the utilization of healthcare resources, performance evaluation, information system strengthening, and internal hospital management to boost technical efficiency. TCM hospitals need to focus on technology innovation, thereby improving the technological progress.

Footnotes

Availability of Data and Materials

Data and materials will be accessible with the permission from the research team.

Declaration of Conflicting Interests

Hao Li is a section editor of INQUIRY. He was not involved in the review of decision related to this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research has been funded by Primary Public Health Services Performance Evaluation in Hubei Province (2020)/grant number: 2305010205.

Ethics Approval and Consent to Participate

Data were all from the official report and did not include the personal information of humans. Therefore, there were no needs to ask for ethics approval and consent to participate.

Consent for Publication

All the authors listed in this paper have already read this manuscript and agreed to submit the paper and for later publication.

ORCID iD

Hao Li

References

China’s Central Party Committee and the State Council. Healthy China 2030 (from vision to action). 2016. https://www.who.int/healthpromotion/conferences/9gchp/healthy-china/en/

Shen

Geng

Shou

Ban

Study on the regional allocation analysis of health resources of Traditional Chinese Medicine Hospitals (in Chinese). Chin J Health Stat. 2018;35(3):445-448.

Gajewski

Kośmider

Nowacka

Puk

Wiciński

Potential of herbal products in prevention and treatment of COVID-19. Literature review. Lit rev Biomed Pharmacother. 2021;143:112150.

Yang

Zhu

Study on regional differences and cooperative development of service efficiency of Chinese medicine hospitals in China (in Chinese). Res Health Econ. 2019;36(12):25-28.

Wang

Suo

, et al. An investigation into traditional Chinese medicine hospitals in China: development trend and medical service innovation. Int J Health Policy Manag. 2017;6(1):19-25.

Hong

Analysis and development prediction of current situation of TCM human resources (in Chinese). Chin J Inf Tradit Chin Med. 2017;25(6):1-5.

Hong

Fang

Cheng

Man

Shi

The status quo and problems of traditional Chinese medicine service system in China (In Chinese). Chin Health Econ. 2010;29(9):3-5.

Ding

Cheng

Fang

Cheng

Shi

Study on the Malmquist Index of operating efficiency in tertiary traditional Chinese medicine hospitals in Beijing (in Chinese). Chin Hosp Manag. 2016;36(7):55-57.

Dong

Zuo

Guo

Liu

(eds.). Data envelopment analysis for relative efficiency measurement of Chinese hospitals: A Systematic Review. Res Health Sci. 2017;2:79.

10.

Masri

Asbu

EZ.

Productivity change of national health systems in the WHO Eastern Mediterranean region: application of DEA-based Malmquist productivity index. Glob Health Res Policy. 2018;3(1):22.

11.

Singh

Bala

Kumar

Janor

Application of DEA-based Malmquist Productivity Index on health care system efficiency of ASEAN countries. Int J Health Plann Manage. 2021;36(4):1236-1250.

12.

Kang

Kraipornsak

Technical and scale efficiency of traditional medicine hospitals and general hospitals in Inner Mongolia, China. Southeast Asian J Econ. 2014;2(1):105-137.

13.

Bai

Man

Hong

Shi

Bian

Do China rural traditional Chinese medicine hospitals provide efficient healthcare to the people? Empirical study from 2013 to 2018 using data envelopment analysis. PLoS One. 2022;17(4):e0267490.

14.

Yue

Zhou

Ning

, et al. Efficiency and productivity of county-level traditional Chinese medicine hospitals in Hubei Province, China: A retrospective study based on 17 years of panel data. Int J Health Plann Manage. 2021;36(4):1308-1325.

15.

Zhu

Shi

Nicholas

Regional disparities in health care resources in traditional Chinese medicine county hospitals in China. PLoS One. 2020;15(1):e0227956.

16.

Simar

Wilson

PW.

Sensitivity analysis of efficiency scores: How to Bootstrap in Nonparametric Frontier Models. Manage Sci. 1998;44(1):49-61.

17.

Simar

Wilson

PW.

A general methodology for bootstrapping in non-parametric frontier models. J Appl Stat. 2000;27(6):779-802.

18.

Liu

Wang

Sara

Yang

Dong

The technical efficiency of community health service centers in Wuhan, China: estimation and policy implications. Inquiry. 2018;55:1-9.

19.

Caves

Christensen

Diewert

WE.

The Economic Theory of index numbers and the measurement of input, output, and productivity. Econometrics. 1982;50(6):1393-1414.

20.

Jiang

Liu

Chen

, et al. The technical efficiency of maternal and child health hospitals in China: a case study of Hubei Province. Reprod Health. 2022;19(1):1-8.

21.

Wilson

PW.

FEAR: A software package for frontier efficiency analysis with R. Socioecon Plann Sci. 2008;42(4):247-254.

22.

Zhu H, Wei L & Niu P. The novel coronavirus outbreak in Wuhan, China. Glob Health Res Policy. 2020;5:6.

23.

Bai

Shi

Cao

, et al. The epidemiological characteristics of deaths with COVID-19 in the early stage of epidemic in Wuhan, China. Glob Health Res Policy. 2020;5:54.

24.

World Health Organization. Listings of WHO’s response to COVID-19. 2020. Accessed November 15, 2022. https://www.who.int/news/item/29-06-2020-covidtimeline.

25.

Liu

Duan

, et al. Prevention and Control of COVID-19 after Resuming General Hospital Functions. Pathogens. 2022;11(4):452.

26.

Zhuang

Cao

Zhao

Han

Yang

The shortage of hospital beds for COVID-19 and non-COVID-19 patients during the lockdown of Wuhan, China. Ann Transl Med. 2021;9(3):200.

27.

Cai

, et al. Assessment of Internet hospitals in China during the COVID-19 pandemic: National cross-sectional Data Analysis Study. J Med Internet Res. 2021;23(1):e21825.

28.

Zhao

Xixin

NIU

Zhou

Tian

The efficiency evaluation of municipal-level traditional Chinese medicine hospitals using data envelopment analysis after the implementation of hierarchical medical treatment policy in Gansu Province, China. Inquiry. 2022;59:469580221095799.

29.

Chen

Cai

Lin

Wang

Impacts of the COVID-19 pandemic on public hospitals of different levels: six-month evidence from Shanghai, China. Risk Manag Healthc Policy. 2021;14:3635-3651.

30.

Kang

Jin

Jiang

, et al. Efficacy and mechanisms of traditional Chinese medicine for COVID-19: a systematic review. Chin Med. 2022;17(1):30.

31.

Wang

Zhou

Wang

Sang

Liu

Current policies and measures on the development of traditional Chinese medicine in China. Pharmacol Res. 2021;163:105187.

32.

Pan

Wenping

Shusong

Kai

(eds.). Evaluation of health resource allocation efficiency of Chinese Medicine Hospital based on data envelopment analysis. Paper presented at: Proceedings of the 2nd International Conference on Social Science, Public Health and Education (SSPHE 2018); January 2019. Atlantis Press.