Trends in Consumption of Quinolones in Chinese Healthcare Institutions: Results of National Procurement Data Analysis (2015–2018)

Chen Chen Xian Jiaotong University Khezar Hayat Xian Jiaotong University: Xi'an Jiaotong University Sen Xu Xian Jiaontg University Ali Hassan Gillani Xian Jiaotong University Wenchen Liu Xi'an Jiaotong University Jie chang Xian Jiaotong University Caijun Yang Xian Jiaotong University Wenjing Ji Xian Jiaotong University Minghuan Jiang Xi'an Jiaotong University Mingyue Zhao Xian Jiaotong university Yu Fang (  yufang@mail.xjtu.edu.cn ) Xi'an Jiaotong University


Introduction
Antimicrobial resistance (AMR) is among the world's leading threats to public health, which has also ampli ed the healthcare cost signi cantly (1). One of the critical drivers of AMR is the overuse of antibiotics. A strong association between AMR and antibiotic consumption has been highlighted by numerous studies (2,3). To cope with this increasing use of antibiotics globally, the World Health Organization (WHO) launched a set of global strategies at the start of the 21st century. According to an estimation, the rate of mortality attributed to AMR could increase from 700,000 in 2015 to 10 million in 2050 without effective control policy (4).
The intensity of the AMR problem is higher among low-and middle-income countries, possibly due to the higher consumption of antibiotics (5). China is among the top countries globally in terms of antibiotic production and consumption, which are widely used in the human and animal sector for the cure, treatment, and prophylaxis of infectious diseases (6). In 2010, China was ranked second in antibiotic consumption by volume in humans (7). A study in 2013 concluded that each of the Chinese residents had used a mean value of 25 grams of antibiotics (8). Surprisingly, the rate of antibiotics prescribing to treat common cold was higher than 80% in China, which should be under 30% as recommended by WHO (9). A signi cant number of factors that trigger the consumption of antibiotics in China include inadequate knowledge of healthcare professionals, patient's pressure to prescribe antibiotics, and nancial compensation for the sale of drugs (10,11).
The rate of AMR is higher in China, owing to a lack of prudent use of antibiotics. A report of the national antimicrobial surveillance system, which used data of 1427 secondary and tertiary hospitals, a rmed the presence of 36% MRSA, 59% third generation, and 54% quinolone-resistant Escherichia coli (12). This increasing rate in AMR compelled the Chinese government to launch several interventions on the prudent use of antibiotics (13,14). In recent health system reforms, the authority to prescribe antibiotics in primary health-care settings was limited to only those drugs that were listed in the National Essential Medicines List (NEML); the sale of these medicines was based on a zero markup policy, thus eliminating the probability of nancial bene t (15). Nowadays, this policy has been implemented in all public sector hospitals. In 2011, a three-year nationwide campaign was launched by the Chinese Ministry of Health to prevent the judicial use of antibiotics, speci cally in secondary and tertiary hospitals (16). In 2012, China's government issued guidelines about the clinical use of antimicrobial drugs to further eradicate the risk of antibiotic sale without prescription (17). There were three categories of antibiotics, including nonrestricted, restricted, and controlled. All antibiotics of the controlled list were removed from the NEML for use in primary healthcare settings. Administrative restrictions were made to prescribe antibiotics from the restricted and controlled category. Various punishments and penalties were imposed for non-compliance (18). In the recent ve-year national action plan, the Chinese government made antibiotics a prescriptiononly medicine by 2020 to combat AMR (19).
There is extensive use of quinolones to treat a large number of infectious diseases, owing to their broadspectrum and optimal pharmacokinetic characteristics. Nevertheless, their widespread and irrational use is also augmenting the risk of bacterial resistance (20,21). Currently, the literature lacks studies designed to determine the trends in antibiotics, speci cally uoroquinolone consumption at a national level by using ATC/DDD methodology, which could frame the picture of antibiotic resistance with higher resolution. Previously, researchers have given more attention to investigating antibiotics prescription changes, coupled with its cost (22,23). Therefore, the current study was designed to measure the trend and pattern of uoroquinolone consumption in China over four years from 2015 to 2018.

Study settings
China is a developing country covering nearly 9,600,000 square kilometers, with a population of 1.428 billion inhabitants (24). In 2019, China's gross domestic product (GDP) was 99 trillion yuan (14.3 trillion USD) (25).

Data collection
As of 2015, the China Drug Supply Information Platform (CDSIP) was launched in China to gather information about drug storage, drug order, and drug delivery. It was made mandatory for healthcare institutions to upload this information regularly. By using CDSIP, the government could update the quantity, price, distribution, and arrangement of drugs. Furthermore, CDSIP helped strengthen the management of drugs. Since 2017, the CDSIP has drug procurement data of 80% of the national healthcare institutions. Therefore, CDSIP is the most comprehensive and reliable data source for national drug procurement data. In this study, we used CDSIP to analyze drug procurement data of 51,935 national healthcare institutions, focusing on determining the trend and pattern of uoroquinolone consumption. For longitudinal analysis, we used procurement data of 41,306 healthcare institutions that had not undergone any change in the four years.
Healthcare institutions were categorized into different types, including primary hospitals, secondary hospitals, tertiary hospitals, upgraded hospitals, urban, and rural primary healthcare centers (PHCs). Various other hospitals included are traditional Chinese hospitals, general hospitals, specialized hospitals, and nursing homes. Community health centers, township health centers, clinics, and village clinical stations were classi ed as PHCs.
The information which was extracted from the database includes the name of the province, the name of the institution, the name of the drug, dosage form, the strength of the drug, pack size, the total number of packs, name of the manufacturer, and the order time.

Data analysis
In the rst step, the procurement data were normalized in accordance with Anatomical Therapeutic Chemical (ATC)/DDD methodology by the WHO Collaborating Center for Drug Statistics Methodology (WHOCC).
We calculated the antibiotic consumption in a de ned daily dose (DDD) as a unit of measurement as per WHO guidelines. DDD is the average daily dose maintained by adults with primary indications; DDDs represent the amount of DDD consumed to measure the frequency of antibiotic use. The formula for DDDs is the number of packages sold × the number of drugs × the strength of each drug/DDD.
Although the CDSIP covers more than 80% of all healthcare institutions in China, the different proportions of healthcare institutions across provincial-level regions and levels of institutions are excluded in this platform owing to a lack of information system infrastructure within those institutions. Hence, the current analysis cannot cover all healthcare institutions in China. We assumed that there is no systematic difference in antibiotic use patterns between the healthcare institutions included and those not included in the CDSIP. We, therefore, used a simple extrapolation to account for the antibiotics consumed by institutions that were excluded from the CDSIP.

Results
A total of 19 speci c chemical substances were identi ed in single or combine antibiotics.

Overall consumption of J01M antibiotics
The use of J01M class of antibiotics was the highest in 2015, with a peak value of 120 million DDDS, however, in the last quarter of 2015, its consumption was reduced to 112 million DDDS. This decline in quinolones' use continues in 2016 with a total consumption of 82 million DDDS in the last quarter. Unfortunately, the use of antibiotics rose abruptly in 2017, with a peak level of 112 million DDDS.
However, there was a marked reduction in consumption in the subsequent year (Fig. 1a). The overall J01M antibiotic use was declined from 25.7-23.5% over four years (2015-2018).
The top seven quinolone antibiotics consumed from 2015 to 2018 were levo oxacin, nor oxacin, moxi oxacin, cipro oxacin, o oxacin, lome oxacin and gati oxacin. Out of total antibiotics, levo oxacin was the most commonly used antibiotic in 2017 with a total share of 61.4%, followed by nor oxacin 27.1% and moxi oxacin 4.1%. Additionally, the use of levo oxacin was continuing to rise from 59.3% in 2015 to 61.3% in 2018 (Fig. 1b). However, the use of cipro oxacin was signi cantly reduced from 4.9% in 2015 to 2.6% in 2018. Complete information about the consumption of all J01M antibiotics is available as a supplementary le.

Consumption and expenditure of J01M antibiotics based on the institution
The use of quinolones was found to be higher in rural PHCs (53.9%) and tertiary hospitals (22.1%) followed by secondary hospitals (14.8%), primary hospitals (13.8%), urban hospitals (6.8%) and ungraded hospitals (0.8%). Despite this, a reduction in overall quinolone consumption was noted in rural PHCs from 58.4% in 2015 to 48.4% in 2018. Surprisingly, there was a marked increase in quinolone consumption in tertiary hospitals from 18.9% in 2015 to 26.4% in 2018. The overall quinolone expenditure was found to be consistently increasing from 21.4% in 2015 to 27.9% in 2018, and the highest expenditure occurred in tertiary care hospitals (56.4%) followed by secondary hospitals (25.6%), and rural PHC (12.8%), (Fig. 2b).
Consumption and expenditure of J01M antibiotics based on the dosage form Oral consumption of quinolones was predominant than the parenteral form in selected hospitals (72.8% vs 27.2%). In 2015, 295 million DDDs of oral quinolones were used; however, this was slightly reduced to 280 million DDDS in 2018, as shown in Fig. 3a. The expenditure of parenteral quinolones was signi cantly higher compared to oral quinolones (83.4% vs 16.6%). However, a decreasing trend was noted in parenteral quinolone consumption from 85.0% in 2015 to 82.2% in 2018 (Fig. 3b).

Discussion
This national-level study is the rst of its kind, which reports on the trend, consumption, and expenditure of quinolone antibiotics over a period in China, providing an opportunity to restructure their policies regarding antibiotic use. Quinolones are broad-spectrum antibiotics widely used to treat multiple infectious diseases due to their optimal pharmacokinetic characteristics. Unfortunately, their irrational use is continuously increasing worldwide, including China; thus, amplifying antibiotic resistance risk (26).  (27). These results could be attributed to the implementation of numerous interventions including strict antibiotic use policies including hospital-based antimicrobial stewardship programs implemented by the Chinese government to restrict the irrational use of antibiotics in hospitals and community (28,29). Besides, in 2012, the Ministry of Health launched Administrative Rules for the Clinical Use of Antibiotics which is playing a pivotal role in minimizing the antibiotic use in different hospital settings as reported in other studies (30). The government has implemented punishments if healthcare professionals and institutions violate the laws and rules of rational antibiotic use, including the loss of accreditation, downgrade in service fees, and dismissal of managers. In severe cases, the doctors could be stopped to prescribe antibiotics, and their medication registration could be revoked. Still, there is a need to do more work to minimize the level of all antibiotics, including quinolones in all hospital settings of China, to meet the recommendations of the WHO. Furthermore, since 2015, a systematic restructuring of public hospitals has been undertaken nationally.
The most integral component of these reforms is the extension of Zero Mark-up Drug Policy (ZMDP) to prevent antibiotic misuse in secondary and tertiary hospitals. ZMDP has clearly shown a reduction in irrational prescribing of antibiotic use as found in various studies (27,31).
Levo oxacin was the most frequently used quinolone antibiotic in our study from 2015-2018 (59.3% in 2015 to 61.3% in 2018). Similar results were found by a previous study where the use of levo oxacin was greater than 30% of the total consumption of quinolones in each year (32). Likewise, levo oxacin was among the top ve most-used antibiotics in a Shandong study (27). This increasing use of levo oxacin has also ampli ed the risk of resistance in different infections, including tuberculosis (33).
In our study, the use of the parenteral form of quinolone antibiotics remains nearly static from 2015 to 2018 (27.9% in 2015, 27.1% in 2018). This slight decline is still far from the standard limits. A high proportion of parenteral quinolone expenditure (more than 80% of the total expenditure in 2018) were found in our study, which is as per previous studies (27). In developing countries, including China, people often prefer the injectable form of antibiotics as they consider them more potent and effective to eradicate their infections. Therefore, some patients force healthcare professionals to prescribe the parenteral form of antibiotics, only (34).
There is a signi cantly higher consumption of quinolones in rural PHCs and tertiary hospitals than in other healthcare settings, which is alarming. A large number of studies have also reported similar results (35,36). The management and use of quinolones in these hospital settings should be controlled by enforcing strict policies launched by the Chinese government. It is advisable that an optimal and effective system should be enforced to determine the irrational use of antibiotics within hospital (37,38). Those antibiotics, used irrationally should be placed on the restricted list, and only written permission should be obtained from the doctors whenever they need such type of antibiotics. This will help in the prudent use of antibiotics within the hospital. Besides, continuous training programs should be launched for all healthcare professionals aimed at the judicial use of antibiotics and antibiotic resistance to equip them with the latest advancements (39,40). To enhance health literacy speci cally about antibiotics, public education campaigns should be instituted regularly to provide awareness of the prudent use of antibiotics (41). This will also help in reducing the pressure of patient demands about prescribing antibiotics.
Certain limitations should be taken into consideration. First, the antibiotic purchase record of the institution is unable to differentiate between inpatient medication and outpatient medication. Therefore, this study could not investigate the use of antibiotics in outpatient and inpatient settings. Second, this study could not nd the antibiotics, which were expired or discarded during storage; likewise, it's di cult to estimate the length of time of storage of antibiotics in the inventory after purchasing by using sales data. Despite the above limitations, this is a pioneer national-level study, which has highlighted the expenditure and consumption of quinolones from 2015 to 2018 in China.

Conclusions
The overall consumption of quinolones was reduced slightly between 2015 and 2018. A decreasing trend in the use of the parenteral form of quinolones and its expenditure was also noted. However, the overall quinolone expenditure is continuously increasing with the parenteral quinolone expenditure greater than 80% of the total expenditure. There is a need to implement a strict antibiotic management system to control the irrational use of antibiotics. Regular education sessions for all healthcare professionals, including physicians, should be arranged.

Declarations
Ethics approval and consent to participate Not applicable.

Availability of data and material
The dataset used in this publication will be available from the corresponding author upon a reasonable request.

Consent for publication
Not applicable.