Serviços customizados
texto em 
Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Citado por Google 
Similares em
SciELO 
Similares em Google 
Permalink
Contribution to scientific literature
Parenteral Nutrient Mixtures (PN) present a risk of microbiological, chemical and physical contamination during their preparation, storage, distribution and administration process; these potential risks must be controlled with high levels of excellence, because their consequences can entail severe damage to patients. Therefore, it is necessary to generate new specific developments that will allow to control PN management and traceability, as well as their constant assessment.
The value of this review consists in showing the first contributions regarding different proposals for the comprehensive management of PN control and assessment, including from graphic notation systems to the most innovative technological developments that can be integrated within clinical practice.
Introduction
Parenteral Nutrient Mixtures (PN) enable to maintain an adequate nutritional status in patients with intravenous administration, when the oral or enteral administrations are insufficient or inadequate in order to supply the essential nutrients. PN can contain up to 50 components with a high potential of physical-chemical interactions between their ingredients, the bag, oxygen, temperature and light. These interactions are potentially iatrogenic, and in some cases can even endanger the life of the patient1.
The great complexity of this formulation and the interdisciplinary nature of nutrition therapy turn this into a procedure particularly prone to errors2. For this reason, PN is considered a high-risk therapy, and can have severe or fatal consequences for patients3. Therefore, it is essential to conduct a normalization or standardization of processes, because when clinical guidelines (practice standards and/or recommendations) are adequately applied by protocol, there is a reduction in their unjustified variability, and this will have a direct impact in quality improvement and the minimization of the risks associated with this type of therapy4.
Given that PN is a process formed by different stages or subprocesses, it is essential to have tools that will allow to assess the adherence in practice to those standards determined during its implementation, as well as to measure the outcomes derived from its introduction.
These tools are also required to allow the follow-up and traceability stated in Law 29/2006, for guarantees and rational use of medications and healthcare products5, and the Good Practice Guidelines for preparation of medications in Hospital Pharmacy Units6, so that, when risks are detected regarding medication quality and safety, it will be possible to react on time and adequately, thus avoiding any damage to the patient.
At the same time, major technological changes have appeared in recent years, which allow the use of computer applications in clinical practice; these will lead to an improvement in quality and higher adherence to guidelines, an optimization in disease monitoring, and a reduction in medications errors7. In this sense, and particularly in the prescription, transcription and preparation stages in the PN setting, efforts have been targeted toward the creation of clinical decision support systems and electronic prescription programs, which act as protocols for clinicians, monitor nutrition therapy, and incorporate alert systems8. Besides, these include systems that will automatically calculate the different components in the mixture, thus reducing the time used in formulation and any potential errors, and also improving patient safety and process quality9.
Therefore, the importance of this review consists in showing the first contributions regarding different proposals for the comprehensive management of PN control and assessment, including from graphic notation systems to the most innovative technological developments that can be integrated within clinical practice and streamline these processes. Consequently, the objective of this article was to review the scientific literature about clinical pharmacy information systems applied to parenteral nutrition management and traceability.
Material and Method
Design
A descriptive study and critical analysis of the articles retrieved in the bibliographic review through a systematic technique.
Data sources
All documents reviewed in this study were obtained through direct on-line search and access to scientific literature indexed in the following databases: MEDLINE (through PubMed); Scopus; Web of Science (ISI-Institute for Scientific Information); The Cochrane Library; International Pharmaceutical Abstracts (IPA) and Academic Google.
Information Treatment
The review included articles published in any country, by any institution or individual researcher, and in any language, that had been published since each of the data-bases questioned became indexed.
In order to define the terms of the search, the Medical Subject Headings (MeSH) was used; this is a thesaurus developed by the U.S. National Library of Medicine. «Parenteral Nutrition» and «Drug Information Services» were considered adequate descriptors (MeSH). These terms were also used, in text format, in order to interrogate the title and abstract record fields (Title/Abstract). The final search equation was developed through the use of boolean connectors to be used in the MEDLINE/ PubMed databases, as appears below:
(“Parenteral Nutrition”(Mesh)) OR “Parenteral Nutrition”(Title/Abstract)) AND (“Drug Information Services” (Mesh)) OR “Drug Information Services”(Title/Abstract)).
It was not necessary to use filters (limits). The same strategy was subsequently adapted to the characteristics of the other databases previously mentioned. The search was conducted from the first available date, according to the characteristics of each database, until May, 2016 (date of the last update).
Article Selection
In order to achieve the maximum level of evidence and the highest number of potential recommendations, it was decided to select for analyses those articles that met the inclusion criteria: the documents should be original and adequate to the objectives of the search, they should not be obsolete (measured by a median date of publication over 10 years)10, and should have been published in peer-reviewed journals. At the same time, the complete text of the article should be accessible. The study excluded those articles about information systems that did not offer data about PN management and traceability; and those articles identified as duplicate publications.
Additionally, as a secondary search and in order to reduce any potential publication bias, there was a review of the bibliographic list of the articles that were selected in the main search, with the object of identifying any studies that were not detected in the bibliographic database review. Specialists in this area of knowledge were also contacted, in order to learn about any potential grey literature existing.
Article selection was conducted independently by two of the authors of this review (JMG and JSV). In order to validate document selection, a standardized methodology was used, that determines that the concordance value between these authors, through the Kappa Index, should be >0.80 (very good strength of agreement)11. Any time this condition was met, any disagreements were solved through consultation with the third author (CWB) and subsequent consensus between all authors.
In order to evaluate the quality of the articles selected, the STROBE (STrengthening the Reporting of Observational studies in Epidemiology) (http://www.strobe-statement.org)12 guidelines for publication of observational studies were used, which contain a list of 22 essential items that must be detailed in the publication of these studies. For each article selected, one point was assigned per each item present (if not applicable, there was no score). When an item included many points, these were evaluated independently, assigning the same value to each of them, and subsequently calculating an average (which was the final result for said item); in such a way that the score of one point per item could not be exceeded in any case.
The continuous control of information validity was conducted through double tables that allowed to detect and correct any errors through a new search in the originals. The studies chosen were grouped with the objective of systematizing the results and making them easier to understand. All relevant data in each article were summarized in a table; specifically, these were coded according to the first author in the bibliographic reference and year of publication, the country where the study was conducted, the population receiving the nutrition, type of nutrition, setting of administration, information system, and variables controlled by the system.
Results
When applying the search criteria described, 153 references in total were located: 24 (15.69%) were retrieved from MEDLINE, 19 (12.42%) from Scopus, 107 (69.93%) from Academic Google, and 3 (1.96%) were contributions by experts. No results were obtained from the search conducted in the Web of Science, Cochrane and IPA databases. From the articles retrieved, 11 (7.19%) were rejected because they were duplicated in more than one bibliographic database.
After applying inclusion and exclusion criteria (see Figure 1), 6 articles were left for review4,13-17, (see Table 1). One article (0.65%)9 was excluded because it dealt with information systems, but did not offer any data about PN management and traceability.
Table 1 Characteristics of the 6 studies (assessed) about information systems for PN management and traceability
PN = Parenteral Nutrition; EN: Enteral Nutrition; TPN: Total Parenteral Nutrition; Hp= Hospitalized patient; Op: Outpatient; SNS= Specialized Nutritional Support.
When assessing agreement among authors regarding its relevancy, this was of 100%.
The six articles chosen presented an obsolescence of 1 year, measured by their median (Burton Kebler Index), with an 83.3% Price Index (proportion of articles with a publication date below 5 years). When assessing the quality of the articles chosen with the STROBE Questionnaire12, the scores ranged between 9 and 17 (median 13.75) (Table 2).
Table 2 Analysis of the Methodological Quality of the studies by the 22 Evaluation Items in the STROBE Guidelines12
a 0 = does not meet the item or any of its parts; 1 = meets the item completely; 0 to 1 = meets partially the item; NA = non-applicable
b = Proportion in which all the items are met, excluding those non-applicable (NA).
All articles originated from Spanish institutions, and were written in Spanish. The year with the highest number of articles was 2015, with three publications (50%). The population of reference (i.e. treated with the type of nutrition on which the information system developed in each article was applied) was heterogeneous: 1 article focused only on hospitalized adult patients17, another one included adult and paediatric hospitalized patients15, 3 articles4,13,16 designed systems that could be applied both to hospitalized patients and outpatients, without differentiating between adult or paediatric patients. And no population of reference was specified in the methodological review by Sanz-Valero et al.14
All articles were applied to PN. Moreover, two of them4,15 developed systems for Specialized Nutrition Support (SNS) including Enteral Nutrition (EN) as well. One of the articles presented data for the use of systems in Clinical Nutrition as well as in Community Nutrition14.
Regarding the information systems detailed in each article, there are differences in terms of the type of system used and the incorporation of new technologies. Three of them4,16,17 used flow charts, or some type of graphic notation, in order to develop their proposals for management and traceability. The rest of the articles were based on computer programs13,14,15.
The oldest study17, published in 2007 under the ISO rules (International Organization for Standardization), developed a normalized system for process description and documentation, with a sequential detail of circuits and associated activities, the staff in charge and guidelines for action. In another study4, a comprehensive dashboard was built to improve SNS practice, based on the identification of monitoring standards and indicators. Regarding the most recent studies, an article also developed a support system for SNS15, but in this case it was a computer application that allowed its normalization from a multidisciplinary approach, and integrated within the electronic clinical record. Another one used normalized graphic notation16, through Business Process Modelling Notation (BPMN), to analyze and describe the different stages within the overall PN process, which facilitated the analysis of each subprocess. The two remaining articles13,14, published in 2016 and 2015, applied new technologies in their approach: in one article, an analysis was conducted on the application of QR (Quick Response) Codes in combination with Web platforms14, while the other one developed a mobile application for Android, which allowed reading different labels (QR, Datamatrix, NFC) and data exchange with a Web server for applying control and traceability procedures13.
However, only two of the selected articles showed the design for a comprehensive management system for PN control and traceability13,16.
The variables controlled by the different systems developed depended on their objective. Thus, those systems created for SNS standardization4,15 included a higher number of clinical variables, from screening and nutritional evaluation of patients, prescription, nutrition formulation, preparation, dispensing and administration to patient monitoring. Additionally, these included management and traceability sections that allowed to evaluate their adequacy to the established standards. Other systems had been exclusively designed for the control of PN management and traceability13,16,17.
Two articles13,16 were based upon Hazard Analysis and Critical Control Point (HACCP) in order to control any potential risks by the processes included in the system.
Two articles4,17 identified and developed quality indicators that controlled the adequate implementation of the established standards. However, these were assessed after the process was completed. The systems designed through computer programs, in two of the articles13,15, allowed to have a record repository that associated and evaluated variables at any point. Besides, these systems allowed the generation of notifications or alerts. On the contrary, two of the articles14,16 did not feature any system assessment method.
Discussion
It could be confirmed that the articles selected for review were completely valid for current times, as well as their interest for the topic chosen: Drug Information Systems applied to PN. In fact, the majority of documents retrieved had been published in recent years. The data obtained (median equal to 1 year) indicate an obsolescence lower than that observed in other articles previously published in the health science setting18. And it appears evident from its documentary analysis that both their obsolescence, measured by their median, and their Price Index, indicate the complete actuality of the matter studied19, belonging to a fully emerging area of knowledge.
When evaluating the quality of the articles through the STROBE Questionnaire12, the score obtained was lower than that of other systematic reviews about observational articles. This might be due to the type of documents selected and the number of items in this form which are not applicable here (NA). Considering the majority of the designs collected in this article, possibly the STROBE Questionnaire might not be the most adequate in order to assess the quality of articles; and that is why the value obtained in the results is not as high as expected from thorough studies. However, there has been an attempt to reduce said limitation, because there is no awareness of the existence of any other questionnaire to evaluate the quality of observational articles, by coupling the absolute values obtained with relative values, and not including those points when their presentation was not applicable20.
Contrary to what was expected, the documents reviewed came from Spanish institutions and were written in Spanish. This goes against usual bibliometric results, which show a clear prevalence of Anglo-Saxon institutions and the English language21. Probably this new line of research observed in Spanish institutions might be caused by the increasing and compelling need to work under quality criteria. There might also be some influence of the Home and Ambulatory Artificial Nutrition (NADYA) Group from the Spanish Society of Parenteral and Enteral Nutrition (SENPE), and the Nutrition Group from the Spanish Society of Hospital Pharmacy (SEFH); both groups have a wide and productive experience in research, and have collaborated in the initiative and creation of the Health Research Project PI13/00464 «Optical recognition technologies applied to the traceability and Hazard Analysis and Critical Control Points (HACCP) of parenteral nutrient mixtures» (http://proyectoqr.umh.es/).
It is a fact that there are articles about information systems for other types of medication, such as biosimilars, where current systems ensure the traceability of biologic products from the manufacturer to Pharmacy Units22. Other articles deal with the processes in the normalization of the supply chain that allow a systematic collection of information23.
The most vulnerable point in traceability compliance is the spontaneous report system, due to the manual nature of data transfer. Any efforts to improve traceability must be focused, at short term, in encouraging health-care professionals and patients to register and report detailed information of the normalized processes. On the other hand, long-term solutions involve a wider access and an increase in electronic data exchange, as was confirmed in the study by Vermeer et al.22.
In the context of clinical nutrition, the information provided by technology will lead to a significant im provement in patient safety. Computer programs are useful tools for prescription validation, calculations, identifying preparations, and process management24,25. This will prevent any errors in transcription, dose calculation and preparation, and will provide information based on clinical recommendations, that will support decision making. There is an increasing number of automatic systems that allow to manage jointly the clinical and lab test follow-up of patients, the prescription transcription and PN preparation, thus improving the quality of the process1. North et al.8, in their review of computer systems for nutrition applied to clinical practice, reached the conclusion that computerized systems improved clinical efficacy, and reduced error rates.
Regarding the setting for applying the different information systems developed in these articles, it stands out that some of them were focused on the hospital setting15,17 and did not monitor any hazards that could occur in home PN, which also requires an adequate strategic plan for intervention, monitoring and follow-up; as well as a management and traceability system to control, for example, transportation, storage and recovery.
In this context, the guarantee for PN quality is based upon the fact that all procedures must be conducted according to normalized protocols; and all procedures conducted throughout the PN process should be recorded, thus facilitating process traceability, and optimizing the supervision of all stages involved. This will allow to evaluate the system as a whole, and to determine to what extent these actions adapt to established standards, with the objective of determining any points with opportunity for improvement and preventing risks. However, those systems designed through flow charts or other graphic notation systems allow evaluation, but only in a retrospective way4,16,17.
One benefit of the systems developed through computer application13,15 is that they can feature a repository of all records, allowing to associate and evaluate variables at any time. And it is worth highlighting the relevancy of using systems based on hazard analysis, which guarantee excellence in terms of control, management and traceability13,16.
Another problem observed is the lack of previous records of incidences and errors (evaluation) at the time of implementing these systems. Therefore, it is not possible to establish a direct comparison or to quantify changes. All studies state the need to conduct subsequent studies to evaluate the implementation of the different information systems.
Additionally, one of the requirements of these systems is that they must be easily implemented with the resources available, at a reasonable cost4. Likewise, it is necessary to train the staff involved, and this must be taken into account in the application of mobile technologies13. For all this, it is essential that all healthcare professionals associated with PN should be committed to this task, integrating it within the culture of the organization, and accepting it as an activity within their daily work23.
The low number of articles reviewed could be considered as a potential limitation for this study. This can be due to the lack of specific Descriptors (MeSH), and as previously mentioned, to the topic actuality, and most of all to the fact that this is an emerging area of knowledge and technological application.
Based on all this, we should reach the conclusion that PN must be integrated within a standardized system, with the aim of ensuring quality and minimizing any risks associated with this therapy. The Hospital Pharmacist will be the key player in terms of adherence to the Good Practice Rules established. Equally, the use of technologies applied to PN would allow to set up management systems which are more complete and easier to apply in a real setting. However, currently this is an untapped potential. Therefore, it is considered necessary to generate new specific studies and developments regarding PN management and traceability, which will ensure their constant monitoring and evaluation.
