In this section, the results obtained through the application of the steps defined in the methods of this work are presented.
3.1 Choice of database to be searched and definition of Query
For the selection of articles, the Web of Science and Scopus database was used because it has data processing and analysis of bibliographic mappings. To complement the study, the Compendex database (Engineering Village platform) was used. The research carried out on the Compendex database did not change the number of articles for the present study.
Definition and application of Query.
For Web of Science
Query: TOPIC: (“Lean Healthcare”) OR TOPIC: (“Lean Hospital”) AND TOPIC: (“Tools”).
Refined by: YEARS OF PUBLICATION: (2020 OR 2016 OR 2019 OR 2015 OR 2018 OR 2014 OR 2017) AND LANGUAGE: (ENGLISH) AND TYPES OF DOCUMENTS: (ARTICLE).
For Scopus
Query: (TITLE-ABS-KEY ( "Lean healthcare" ) OR TITLE-ABS-KEY ( "Lean hospital" ) AND TITLE-ABS-KEY ( "tools" ) ) AND ( LIMIT-TO ( PUBYEAR , 2020 ) OR LIMIT-TO ( PUBYEAR , 2019 ) OR LIMIT-TO ( PUBYEAR , 2018 ) OR LIMIT-TO ( PUBYEAR , 2017 ) OR LIMIT-TO ( PUBYEAR , 2016 ) OR LIMIT-TO ( PUBYEAR , 2015 ) OR LIMIT-TO ( PUBYEAR , 2014 ) ) AND ( LIMIT-TO ( DOCTYPE , "ar" ) ) AND ( LIMIT-TO ( LANGUAGE , "English" ) ).
3.2 Selection of articles
In the selection criteria, the search string “Lean healthcare” OR “Lean hospital” AND “Tools” was used. This keyword choice represents most of the terms described within this literature. With this search criterion, in the database Web of Science were found eighty-six publications, fifty-six publications were found in Scopus and twenty publications were found in Compendex, identifying a total of one hundred and sixty-two articles. The next step was the selection for the year of publication of the articles. Only publications made between 2014 to April 20, 2020 were considered. In this stage, the total of seventy-eight publications resulted in the Web of Science, in Scopus there were forty-one documents and in Compendex there were twelve documents.
In addition, three filters were defined as criteria for selecting documents. The first filter selects only documents of type “Article”. In this step, thirty-two records were excluded in the Web of Science database, in Scopus twenty-two records were excluded and in Compendex ten records were excluded from the selection process. With forty-six articles selected on the Web of Science, nineteen articles selected on Scopus and two articles selected on Compendex, the second selection filter was English language. In this step, two articles written in German and three articles written in Spanish on the Web of Science were excluded. Based on these criteria, forty-one articles were selected for the work by the Web of Science database. One article written in German and two articles written in Spanish were excluded from the Scopus database. Based on these criteria, seventeen articles were selected for the work of the Scopus database. In the Compendex database, articles were not excluded, because two articles selected in the previous step were written in English. Based on these criteria, two articles were selected for the work of the Compendex database. In the eligibility stage, she obtained a total of sixty records identified in English.
Replicated records were excluded from the selected studies in a total of twelve replicated articles. Therefore, forty-eight documents were included for the analysis, following the criteria: applied research methodology, the mentioned country, the health area covered by the document, in addition to the Lean methods and tools used and the results of each study, classifying them.
Thus, a meta-analysis was used, addressing the PRISMA method according to Moher et al., (2009) for the selection of the analysed articles. Also addressed by Lobo et al., (2018). Being illustrated in the flowchart represented in figure 2.
3.3 Reading and analysis of texts
3.3.1 Implementation areas
In this step, the analysis of the articles selected for the study was carried out, with the areas and departments of health that applied the methods and tools of Lean Healthcare, as shown in Figure 3. The applicability preference of the methods and tools can be highlighted in hospitals as a whole, found in fourteen documents, emergencies and the surgery department, found in six documents. Even so, it is possible to verify the success of the tools application in other health sectors such as oncology, mental health center, cardiology, radiography, pharmacy, public hospitals, health organizations, chemotherapy, urology, Intensive Care Unit - ICU and among other sectors, however, they were found in fewer documents in the literature.
Table 2 compiles the results found in this section, relating the areas of health with the respective article, the number of times that each area has been addressed in the literature and the percent of each health area actuation in analysed cases.
3.3.2 Lean methods and tools mentioned.
The other stage of the study was to identify, among the forty-eight documents selected, the methods and tools of Lean that are being applied in health departments identified in the previous stage and their purposes of use. Thus, it was found out which and how these methods and tools were being implemented and the results caused by their application. Figure 4 shows the percent of times that each Lean method and tool was addressed in the selected studies through of the methods described in item 3.2. The use of Value Stream Mapping to find problems such as bottlenecks and delays in production processes, helping to solve problems in different health sectors, is cited in twenty-four articles selected in this research, with seventeen percent of the total number of lean tools applications, appointing a preference of use.
Another tool addressed as preferential is the 5S, being used in ten percent of the studied applications and being presented in fourteen documents, and the Kaizen method with the application being cited in thirteen articles and used in nine percent of studied applications, the 5S acts, above all, in the improvement of quality, using 5 “senses”, sense of use and discard, sense of order and organization, sense of cleanliness, sense of standardization and sense of self-discipline. Its importance in health can be attributed to the need for a clean and sterile environment. According to Kanamori et al., (2015) there is a reduction in the time to search for items; greater capacity of the team to move around the office; centrality of the patient; reduction of waiting time for patients; better guidance for patients; improved sterilization processes and others. These were the results found by applying the 5S tools and Kaizen methods.
Following by the tools Visual Management, Standardization and Kanban, in this order with ten, seven and six articles mentioning their use. It was also observed in the literature the application of Lean tools along with other methods such as DMAIC, found five times, Analysis of variance (ANOVA), being mentioned twice, Simulation, found three times, Total Quality Management (TQM) and Total Quality Control (TQC), found twice each, Theory of constraints (TOC) and Fast-Track, each being found once in the literature. These results are best seen in figure 4.
3.3.3 Country of application
Figure 5 shows the percentage of publications by countries. It can be seen that Brazil and the United States are two countries that stand out the most with the largest number of publications related to the theme Lean Healthcare. Brazil has thirteen publications found within the period studied and in the United States with six publications, which are the two main countries in relation to the number of publications, citing the use of tools and the acceptable results of the applications.
We also highlight Sweden with five, Italy and England with four publications each, Ireland and India with two publications each. Scotland, Canada, Spain, Thailand, Kuwait, Netherlands, New Zealand, Oman, Czech Republic, Senegal, Turkey, Hungary, Lebanon and United Arab Emirates published one article each. In two articles studied - more than one country was presented, such as the article by Leite et al., (2019), had its study done by the University of London, England, however, the analysed case study was done in Brazil in an emergency of the Unified Health System. Alnajem et al., (2019) Literature Review the from 2013 to 2018, conducted by the Gulf University for Science and Technology in Kuwait, University of Derby and Heriot-Watt University, England.
3.3.4 Main journals and authors
Among the articles studied, the following Journals with the largest number of publications found in this research were highlighted: JOURNAL OF HEALTH ORGA-NIZATION AND MANAGEMENT, PRODUCTION PLANNING AND CON-TROL, BMC HEALTH SERVICES RESEARCH, BENCHMARKING: AN INTER-NATIONAL JOURNAL, with five, four, three and two publications in this order with the theme Lean Healthcare. Such Journals have together fourteen publications used to build the present study. Table 2 shows the main Journals found in this study and the respective number of publications. All other unlisted Journals have one publication each, about theme Lean Healthcare.
Among the authors studied are: Tony Butterworth, from the University of Lincoln, England; John SG Wells, Waterford Institute of Technology, Ireland; in addition to Mark White, from HSE-South, Ireland; Moacir Godinho Filho, Federal University of São Carlos - UFSCAR, Brazil; Jacopo Guercini, University Hospital of Siena, Siena, Italy; Federico Barnabè, University of Siena, Siena, Italy; Maria Cleofe Giorgino, University of Milano-Bicocca, Milan, Italy. The seven authors are responsible for at least three publications each, of the documents analysed in this research. Table 3 shows the highlighted authors and the respective number of publications, with three or more publications.
3.3.5 Ways to implementation and their results
Analysing the selected scientific documents, we found the preference for the applicability of the Value Stream Mapping tool in several areas of health following the 5S tools and Kaizen method, which is one of the objectives of the article to identify what Lean tools are most used in Lean implementation through different areas of health.
Thus, it was observed the primordial ways of implementing of tools and the main results obtained. In general, as a form of implementation, there are team training workshops, project simulation, meetings with the team, information gathering, separation of activities that generate value and those that do not generate value, engagement between employees and questionnaire with patients to identify what is value to the patient and what the patient identify as waste. Thus, the fundamental results were shown: decrease in hospital waste such as out-of-date medicines, decrease in stock, patient satisfaction, decrease in lead time, stabilization of the process, increase in the number of visits, decrease in errors doctors and errors with blood collection, reduction in the average time for exams delivery, reduction in patient's stay time, cost reduction in the appointment and decrease in the number of appointment and exam cancellations. Therefore, some of the studies analysed were highlighted below, highlighting the ways to implement the Lean tools and the results obtained. The following are related thirteen cases that show the application of Lean tools in health services and its achievements.
The tools of Lean 5S and Kaizen were implemented in the sterilization of surgical materials in Brazil in the study by Fogliatto et al., (2019). For implementation, Lean principles were studied, and it was divided which surgical materials could be prioritized and which could be rationalized to reduce costs and the burden of sterilization processes, divided the teams into Kaizens groups to achieve the objective of continuous improvement. They reported the results proposing a strategy to address the rationalization of trays, prioritizing complex and frequent surgical specialties, and using group technology to agroup surgical trays, streamlining specialized analyses. Thus, they reduced the number of instruments by an average of 9.75% and the time to assemble trays by 9.68%, achieving an annual savings of US $ 285,756.00 in sterilization processing costs.
In the study by Barnabè et al., (2018) he applied the principles of Lean thinking and shared competences in collaborative processes in a hospital, devoted to educational tools for training health decision-making agents. He used the role-playing game (RPG) as a simulation measure and team training. It guaranteed the satisfaction of patient demand, patients reported improvement in the process, improvement in lead time and quality in the process according to health professionals and patients.
Barnabè et al., (2017) proposed a simulation game that not only provides a suitable physical environment for the purpose (an open space within the hospital), but also stimulates and encourages the use of a large number of medical instruments (for example, a portable chest X-ray machine), administrative documents (for example, medical records and laboratory tests) and Lean tools to make the simulation as real and engaging as possible. The game challenges participants with work-related tasks, a variety of possible scenarios (for example, a growing patient demand) and a complete set of metrics to measure their performance. In particular, a dedicated business intelligence software program is used to monitor and evaluate performance. The graphs, key performance assessment, tables and panels inform the discussion and decision making during all phases of the game. As a result, participants obtained recognition of the simulated environment as realistic (58.8%) or absolutely realistic (17.6%). In total, 86.7% of the participants perceived an increase in knowledge and understanding about the specific simulated process. 93.4% of the participants emphasized that the basis of a simulation game was more useful and effective than traditional classroom training. Specifically, players stated that this simulation game encouraged them to reflect and use / implement various Lean tools and techniques (for example: 5S technique, visual management tools, various Lean metrics, value stream mapping) during the simulation. Overall, 94% of the participants were willing to participate more often in the simulation game.
Barnabè et al., (2019) Identified two streams: of knowledge and practical. In his study, Lean training was reported to identify value creation for decision makers have defined a measurement framework to identify what is of value to the patient and training of staff for Lean implementation. Results obtained were improvement in lead time, reduction in the displacement of people, reduction in the cost of processes, reduction of walking time to 154 km, quality of care and safety in health processes.
Boronat et al (2017) used three steps for the implementation 1) team training and improved feedback among professionals, 2) process management and super-specialization and 3) improvement of assessments (continuous improvement). The assessments were obtained from the hospital's information systems. The main source of information was the Balanced Scorecard for health systems management. The comparison with other autonomous and national urology departments was carried out through a platform, with the help of the Hospital's records department. A baseline was established with the assessment obtained in 2011 for the comparative analysis of the results after the Lean Healthcare implementation approach. The results were: high professional satisfaction, improving quality assessment, reaching a risk-adjusted complication rate of 0.59 and a risk-adjusted mortality rate of 0.24 in 4 years. A value of 0.61 was reached with the efficiency assessment, with savings of 2869 stays compared to the national benchmarking. The risk-adjusted readmissions index was the only assessment above the standard, with a value of 1.36, but with a progressive annual improvement.
In the article by Cheng et al., (2015), were held Lean training workshops, a collection of information to establish the current status in each location of the mental health center. The results mentioned in this study were the stabilization of the operational processes in current state within each of the teams of the mental health center, identification of problems and identification of operational issues.
In the study by Demosthenes et al., (2015) for implementing Lean identified items with high cost for which there were less expensive alternatives, but with effective characteristics. As an educational intervention, were carried out the main rounds revising the research results, as well as recommendations that would allow changes in health care. In total, 50 of the 70 suppliers (71%) answered to the survey. Vaginal insertions of hydrochloride acetate / pramoxin hydrochloride and dinoprostone were the target of intervention. The use of vaginal dinoprostone insertion decreased by 50.5%, with savings of US $ 66,500 when comparing the pre-intervention period with the post-intervention period. The use of hydrocortisone acetate / pramoxin hydrochloride decreased by 90%, with savings of $ 92,655. Combined, the decrease in the use of these products led to savings of US $ 159,155 within a year after the intervention. Through the use of research and educational intervention, they demonstrated that simple interventions can lead to changes in Lean healthcare quality.
In the studies by DiGioia et al., (2015) and Efe and Efe (2016) they used similar means for the implementation of Lean and obtained similar results. As implementation mean, they used employee engagement and training, space renovation, analysis of patient flow and process redesign. As a result, the elimination of waste, patient satisfaction, decreased expenditure per patient and length of stay were reported.
Gupta et al., (2018) with their article showed the use of VSM, Pareto chart and Ishikawa diagram (or fishbone). The application of these tools helped to identify the use of other Lean tools such as 5S and Visual Management. For that, they had weekly meetings to discuss progress, training employees and identification waste. The results were reduction in the response time for clinical examinations, patient satisfaction and teamwork.
In the work of Haddad et al., (2016) to comply with the objective of improving flow, reducing time and changing culture, the types of waste in each process were identified, making it possible to build a table describing the phase, the process, the description of the waste and the type of waste. The construction of a simulation model with the support of the Arena software, which identified the move time in the hospitalization sector. As a result, the patient's total time in the admission system was improved by 43%, increase of 95% in the patients' confidence level and the processes became standardized.
With the research by Costa et al., (2015) they discovered problems and opportunities through the application of VSM and DMAIC in five health departments where they analysed patient flows, exams need and patients' schedule. They obtained reduction in lead time and financial costs, 78% reduction in costs with pharmacy, 42% reduction in waiting time reported by patients and 93% reduction in blood analysis time.
Romano et al., (2015) in the emergency department of a Hospital analysed the use of the Fast-Track tool. Through medical screening, patients are sent to Fast-Track using the Emergency Severity Index parameter. Fast-Track has access to those less serious patients for whom the system's crossing time is supposed to be shorter. In medical screening, the patient is separated by colors, red (patients with a critical condition), yellow (patients with a potentially critical condition), green (patients who need a medical service) and white (patients who do not have changes in vital functions). Patients are sent to a specific emergency area according to their classification, the shock room receives all patients identified with the color red, in this environment it has a multidisciplinary team and all the necessary equipment to save lives. In the Urgency area, patients identified with the yellow and green code are received, the distinguishing characteristics of this area is the system's flexibility in accepting additional patient flows (agglomeration). The area of minor codes is dedicated to the evaluation and treatment of patients with minor problems. Assisted waiting area for patients who need to complete treatment but are unable from accessing the hospital due overcrowding. Brief observation, for patients that it is not possible to make a decision about the result in the first hours of arrival at the emergency room. Finally, a short intensive observation area, patients with trauma and toxicology, in which the diagnostic procedures are not exhaustible in a few hours. In this way, they improved the flow of patients, allowing them to assist almost everyone in the ward and reallocating resources in a timely manner.
In this way, the Lean approach such as VSM, 5S, Kaizen and other tools mentioned in this work, can be applied to different sectors in health. Such as clinics, hospitals, surgical center, oncology, geriatrics, gynecology and obstetrics, neurology, emergencies, cardiology, Intensive Care Unit, hospital pharmacy, SUS, mental health service, chemotherapy, radiotherapy, sterilization, maternity and primary care. The use of such methodologies is recommended for health departments that wish to achieve better results and efficiency during health procedures. It is worth mentioning two related cases, one by Flogliatto et al., (2019) and Demosthenes et al., (2015), which presented results of annual savings of US $ 285,756 in sterilization processing costs and US $ 159,155 using educational research, which together generated an annual savings of US $ 444,911, resulting from the implementation of Lean Healthcare.
3.3.6 Barriers and opportunities found
In this stage of the study, it focused on finding in the studied documents the barriers and opportunities to start the Lean journey in health services, mentioned in this work. Therefore, some of the studies analysed were highlighted below, highlighting the barriers and opportunities to implement Lean tools in health services and the results obtained.
The application mentioned by Leite et al., (2019) reports the use of Lean tools in an emergency sector agreed by the SUS in Brazil, aimed at identifying and understanding the barriers to implementation of Lean tools in the public health system. The identified barriers could be answered with Lean management consultancy, six types of barriers were discoveries: 1) Influence of doctors in the process, 2) Patient behavior, 3) Restrictions related to resource management affecting the clinical team, 4) Impact of the SUS model on medical work, 5) The model that SUS operates creates restrictions, 6) Influence of team behavior as a barrier to Lean implementation . The results showed that there are two types of barriers to Lean implementation in health area: ostensible and underlying. Ostensible barriers, in general, are common during the Lean journey, but with deeper causes that influence its creation. These barriers come from the literature and knowledge of professionals. On the contrary, underlying barriers are based on rich qualitative data that emerge as the root cause of the ostensible.
Drotz and Poksinska (2014) reported the introduction of Lean culture and leadership through daily meetings, teamwork, through Value Stream Mapping, it is possible to see opportunities for implementing other tools such as Kaizen, visual control, 5S, takt time, work standardization and Poka-Yoke. He mapped risks and safety and analysed what creates value for the patient. Reported results were the reduction of waste, greater interaction and cooperation between employees and patients.
Eiro and Torres (2015) also identified that through the Value Stream Mapping there’s opportunity to implement other Lean tools such as PDCA (Plan, Do, Check and Act), A3, FMEA (Failure Mode and Effect Analysis), Kaizen, 5S, work standardization, Total Quality Management and Visual Management. Therefore, its mapped risks and safety, and analysed what generates value for the patient. Results: reduction in lead time, reduction of errors with blood collection, movement of such materials and in the performance of the exam, interaction between nursing and pharmacy which allowed less waste with medicines and quick search for specific medicines, decreased movement of health professionals.
New studies can be developed regarding specific barriers for adopting the Lean approach in health services. In addition, it is possible to study how other sectors such as urology, chemotherapy, maternity and primary health care are seldom mentioned for the applicability of Lean methods and tools in health services.
3.4 Statistical and graphic analysis of the selected records
In the identification stage of the articles it resulted in 48 documents with the union of the Web of Science, Scopus and Compendex databases. Figure 6 shows the graph of distribution of articles over the period from 2014 to April 20, 2020, illustrating the number of these publications per year.
Analysing the number of publications per year, since 2014 there a growing trend, with a peak of publications in 2015 and 2017. This indicates that the academic society's interest in this topic has increased and is increasingly discussed by professionals and researchers until the year 2017. In addition, it is noted that the number of publications fluctuates over the years and between the years 2018 to 20 April 2020 the number of publications with the theme about Lean Healthcare decreases. In 2017 there was a greater interest in publications on Lean Healthcare.
The last step of this SLR was to generate, through the VOS Viewer software, the co-citation and co-occurrence networks of keywords, for that it was necessary to join the bibliographic data acquired in the Web of Science and Scopus databases.
Co-citation map.
The co-citation network, shown in figure 7, is composed of 8 nodes and 28 edges. The size of the nodes indicates how many times they were mentioned by other authors. Among them they are mentioned in the same way, because the nodes have the same size.
Keywords co-occurrence maps.
In figure 8, it is about the keyword co-occurrence map made in VOSViewer software, to address the keywords used in the articles. Note that the possible Hot Topics are: “Lean healthcare”, “quality improvement”, “human” and “total quality management”. The least used keywords are: “Hospital operations”, “person-centered care” “Management” and “lean thinking”. The keyword co-occurrence network, illustrated in figure 8, is composed of 238 nodes and 2406 edges. The size of the nodes indicates how many times they were used in the articles. The generated network obtained 16 clusters, that is, 16 sets of items included in the map. Therefore, it was visualized, highlighted in the keyword “human”, it has been mentioned in the literature more over the years, due to the preference of humanized treatment focused on the client / patient defined as one of the pillars of Lean (respect and transparency) which is linked to the National Humanization Policy (PNH) created in 2003 linked to the Ministry of Health in Brazil.
Publication bias
In the present study there was no publication bias because studies were equally studied within PRISMA as shown on figure 2.