The pre-analytical process management status and influencing factors of laboratory test before prescribing antimicrobial in developing country

This cross-sectional survey was a self-administered questionnaire distributed to all secondary and tertiary hospital departments related to PP (including collection department, specimen transport department, inspection department) and several clinical departments with a high frequency of antimicrobial use (respiratory medicine, urology, ICU, neurology, endocrinology, and orthopedics, etc.) in Hubei, China. The area where the questionnaire was sent in this study is consistent with the distribution of medical and health resources in Hubei Province. On the whole, the six cities of Wuhan, Huangshi, Shiyan, Yichang, Enshi and Xiangyang are the areas with the highest amount of medical and health resources. And the facilities we surveyed covered these areas. This survey was implemented in June 2, 2021 and the questionnaires were recalled after 1 week with a reminder. According to the degree of participation in the pre-analysis process and the ratio of human resources of each hospital, one director of the infection control department, at least two staff of the transportation department and the inspection department, at least five clinicians of key departments and eight nurses of each hospital were required to fill in the questionnaire. Relying on the Hubei Provincial Infection Management and Control Center, we distributed questionnaires to all hospitals under its jurisdiction, with a response rate of 100%. After screening, 109 hospitals’ responses met the quality requirements of this study and were included in the analysis. There are 109 infection control managers, 2539 clinicians, 4215 nurses, 980 specimen transporters and 266 specimen inspectors.

To prepare the questionnaire, this study reviewed relevant articles on the practice and impact assessment of antimicrobial management plans at both the institutional and individual levels. Based on Specimen collection and transport in clinical microbiology by the National Health Commission of the People’s Republic of China in 2018 and the 2015 edition of the Guiding Principles of Clinical Application of antibiotic, the issues of supervision are designed from the perspective of management or policy implementation. And then follow-ups with the evaluation by expert judges or panels. Experts involved in the study included five hospital infection prevention and control staff, five clinicians and nurses in key departments, and five relevant personnel involved in the inspection. We used the content validity ratio (CVR) to evaluate the content validity of each item in the questionnaire. Refer to the minimum acceptable value of the content validity ratio reported by Lawshe, the evaluation results of 15 experts are greater than the acceptable minimum value of 0.49, indicating that the content validity of this questionnaire is good. Further details of the research design can be obtained at the hospital infection center of Hubei Province (whcdc. ORG). Eventually, the questionnaire was divided into five parts by different survey departments and subjects.

Almost every part involved the investigation of demographic characteristics, the status of regulatory systems related to PP, performance appraisal, publicity and education, etc. In addition, we designed management evaluation indicators for very department (request rate of non-restricted use of antimicrobial (NR), request rate of restricted use of antimicrobial (RR), request rate of special-grade use of antimicrobial (SR), zero obstacle rate of specimen collection (ZR), zero obstacle rate of specimen transportation (TR), zero obstacle rate of specimen inspection feedback (FR)) to reflect the situation of hospital supervision of PP. The specific definitions and calculation formulas were displayed in Table 1. These indicators were designed to evaluate the management impact under the best conditions of PP.

Questionnaire was sent to the infection control department of each hospital by e-mail. The person in charge distributed it to each part, collected it after 1 week and conducted quality audits, then returned it by email in a specific format. The questionnaire was coded according to medical institutions for summary and sorting. EpiData is used for data entry of paper questionnaire. In order to reduce the bias in the investigation process, we have carried out the following work: 1) Firstly, we held a meeting with each IPC manager before investigation. We fully informed them of the survey objectives and specific content at the meeting and informed them of the specific requirements for distribution, collection and audit of questionnaires. 2) We created a review section on questionnaire requiring the person in charge to check and sign when the questionnaire is received. 3) Our research team arranged specially-assigned persons to liaise with each hospital and the problems faced by IPC managers can be resolved in a timely manner. If necessary, we will go offline for guidance. 4) During the investigation, the IPC managers sent the recycled questionnaire to us via email in the form of electronic scanning every day. Team members will further audit to ensure that the respondents meet the requirements.

Descriptive statistical analysis is used to analyze the different distribution of hospital department characteristics and management factors. Fractional probit regression model was applied to analyze the influencing factors because the values of dependent variables were between zero and one and may also be equal to zero or one. Fractional probit regression model is an extension of generalized linear model (GLM) [23]. It is a kind of functional forms, which avoids most known issues of traditional econometric models for bounded variables [24]. Fractional response estimators fit models on continuous zero to one data using probit, logit, heteroskedastic probit, and beta regression. Probit regression can be used when the endpoints zero and one are included. The log-likelihood function for fractional models is of the form:where N is the sample size, yj is the indicator, wj denotes the optional weights, lnL is maximiz, where xj are the covariates for individual j, zj are the covariates used to model the variance of the outcome for the heteroskedastic probit model, and Φ is the standard normal cumulative density function.

Among the 109 hospitals surveyed, there are 60 secondary hospitals and 49 tertiary hospitals. The hospitals that responded covered the area with the highest amount of medical and health resources, such as Wuhan, Huangshi, Shiyan, Yichang, Enshi and Xiangyang and so on. The demographic characteristics of each department were shown in Table 2. PP related management team members have a single professional background, mostly nursing (88.1%). More than half of the respondents from the clinical department (secondary hospital: 63.4%; tertiary hospital: 60.8%) and the collection department (secondary hospital: 63.4%; tertiary hospital: 60.8%) claimed that the performance appraisal frequency was once a month, and nearly half of the respondents from the inspection department (secondary hospital: 46.7%; tertiary hospital: 44.2%) said that there was no appraisal. However, there was no appraisal related to PP in the transportation department. For training and publicity, the proportion of respondents who participated once a year in all departments was higher. More than 90% of the collection and inspection departments were supported by related guidelines. Information systems were used to support inspection departments (93.2%) more than clinical (85.6%) and collection department (81.7%) (Table 3).

There was a significant difference in the PP before prescribing antimicrobial of clinicians in secondary and tertiary hospital before the use of non-restricted use (X² = 22.968; P < 0.001), restricted use (X² = 29.466; P < 0.001) and special-grade use antimicrobial (X² = 27.317; P < 0.001). The proportion of that in tertiary hospitals was higher when prescribing non-restricted antimicrobial, while that in secondary hospitals was higher when prescribing restricted and special-grade antimicrobial (Table 4). And there was a significant difference also in the collection (X² = 41.256; P < 0.001), transportation (X² = 29.835; P < 0.001) and inspection (X² = 6.514; P = 0.038) departments between the two levels of hospital. The rate of zero obstacle of collection was 27.3% and of transportation was 19.4%. Nurses in the collection department mainly encountered obstacles such as bacterial infection at the collection site (34.2%) and insufficient sample collection volume (39.3%). In the process of transporting samples, too many samples (51.3%) and long delivery time (30.4%) were the interference obstacles with more feedback from the transportation staff. For inspectors, lack of timely feedback channels was the mainly obstacles (38.3%) (Table 5).

According to the analysis of the fractional probit regression model, we found that for all departments, the provision of guidance and training had a significant positive impact on PP. Low frequency performance appraisals had a positive impact on clinical departments and collection departments. However, the publicity of PP had no significant positive impact on most indicators.

Specifically, for the PP of three types of antimicrobial, training had a positive and significant impact on it. The monthly performance appraisals frequency also had a positive impact, and only high-frequency appraisal had a positive impact on the request rate of restricted and special-use antimicrobial. In addition, information system support had a significant positive impact on the inspection of non-restricted and restricted types of antimicrobial. It can be explained that clinicians’ interpretation of the test results before the use of restricted and non restricted antimicrobial needs more information technology based auxiliary support.

For the collection department, providing guidance and training had a positive impact on nurses’ zero barrier of collection, and compared with low-frequency assessment had a positive impact. Publicity twice a year also had a positive impact, while the frequency of four or more times a year had a negative impact. It can be found that low-frequency performance appraisal and publicity helped to promote zero obstacles in the collection process. For the transportation department, training had a positive impact on the zero interference rate, while compared with no publicity, three times a year had a negative impact on that. For the laboratory department, performance appraisal, training and publicity all had a significant positive impact on promoting the problem feedback and communication between the inspectors and the clinical department and the collection department (Table 6).

From the organizational level, the antibacterial drug management plan was carried out by the infection management department in most hospitals. However, we found that the professional background of the organization team for infection management in the hospital is single, mainly nurses. Since a pre-prescription antibiotic inspection requires the cooperation of multiple departments, the management team should achieve a cross-professional combination to improve inspection efficiency and value. The diagnostic management plan proposes that laboratories working with the management team can reduce unnecessary tests and false positive results and better improve patient care [25]. Knobloch et al., suggested that we need to adjust the management plan in combination with the local need and influencing factors [26].

The antimicrobial stewardship program usually focuses on education interventions in many countries [27, 28]. Studies on improving the quality of hospital infection prevention and control have proposed to strengthen training, publicity and performance appraisal [26, 29]. However, our research found that different frequencies of training, performance appraisal and publicity have different effects on the quality of the pre-analytical process, and also have different effects in different departments.

For clinicians in key departments, although the request rates of different types of antimicrobial basically met the requirements, we found that there were differences between different levels of hospitals, and the impact of management factors on them was different. Different from the other two types of antimicrobial, only low-frequency performance appraisal has a significant impact on non restricted-use antimicrobial. Non restricted-use antimicrobial were usually provided in empirical drug use schemes, and the demand for microbiological test results is not very necessary, so the management requirements are relatively loose. This finding suggests that for most clinicians, an overly frequent review regime may be detrimental to the promotion of PP on non-restricted antimicrobial. Some studies have noted that employees’ perceptions of fairness in performance appraisal also impact employee performance, including perceptions of the rationality of the evaluation cycle. In addition, the rationality of performance evaluation content for different levels of employees also affects their performance [21]. We also found that the hospital’s publicity activities on clinicians did not have a positive impact, and even had a negative impact in collection department and transportation Department. It is understood that the PP publicity in the survey area is usually distributed in the form of brochures. This single form is mostly considered ineffective. For the specimen collection department, it can be seen that providing a standard operation guide and training is very important to reduce errors in the collection process, and our results were in line with previous studies [29]. At the same time, we are also concerned that the improvement of collection equipment and technology is also an urgent need to improve the quality of PP. It is suggested that hospital managers need to pay attention to what is the real management demand to improve the inspection work, and the technical and awareness intervention measures of different departments related to the inspection should be different in terms of methods and degrees [14]. For the transportation department, we found that due to a lack of clear regulations regarding the establishment and ownership of the department, many hospitals lack targeted supervision systems. The problems in specimen transport were not resolved in time, which became the most easily neglected link affecting the quality of PP. For the inspection department, we found that the provision of guidance, training and appraosals had a significant positive impact on PP. However the inspection department had fewer performance appraisals related to PP and few laboratory communicate with clinical or other departments during the PP. The pre- and post- analysis phases were considered to be the stage with high incidence of laboratory inspection errors and lack of attention. Researchers emphasized that it is necessary to improve the laboratory personnel’s attention to the sample quality in the PP and the interpretation and feedback of the results after analysis [24]. Therefore, it is necessary for managers to strengthen the appraisals of inspectors in the above process.

In addition, we found the positive impact of information technology on improving the quality of PP. Many studies have proposed that the first way to improve the nursing quality and safety of inpatients is to use a well-designed information system to prevent and manage laboratory errors, and promote the continuous exchange of information among doctors, nurses and laboratory experts [30, 31]. Our results are consistent with them. Compared with regular education and training, information system can provide more timely decision-making guidance. A study from Portugal showed that good communication between doctors and microbiology laboratories should be considered a priority for any ASP [11]. Efforts should be made to improve data sharing between doctors and microbiological laboratories, such as the provision of antibiotic prescription guidelines suitable for local hospital epidemiology, and easy access to hospital antibiotic sensitivity models and epidemiological databases. Multiple Department roles are designed in the inspection process. Good communication can reduce the risk of errors before laboratory analysis. As a timely tool, information technology can promote multi-party information sharing and communication. However, at present, in most of the tertiary and all secondary medical institutions, there is no perfect information system to support the inspection, and the functional requirements of the information system need to be further collected in various departments.

A limitation of the study was that the content of the survey was not comprehensive enough. This could have introduced some bias and omit some possible influencing factors, however the relatively professionals involved in the management of inspection request, providing advice from real practice for survey item, could have reduced this risk. The distribution of the survey may be another limitation. The number of departments surveyed was limited and may not be representative of the wider hospital in this country. Since our survey is the first baseline survey in China, the selected hospitals in a central province are enough to reflect the current average level of examination.