Introduction: Medication errors are a major cause of illness and hospitalization of patients throughout the world. This study examines the situation regarding medication errors in the Armed Forces Hospital, Kuwait since no literature exists of any such studies for this country. Several types of potential errors were studied by physicians, nurses and pharmacists. Their attitudes to the commission of errors and possible consequences were surveyed using questionnaires. Additionally, patient medical records were reviewed for possible errors arising from such actions such as the co-administration of interacting drugs. Methods: This study included direct observations of physicians during the prescribing process, pharmacists while they dispensed medications and nurses as they distributed and administered drugs to patients. Data were collected and compiled on Microsoft Excel spreadsheet and analyses were performed using SPSS. Where applicable, results were reported as counts and/ or percentages of error rates. Nurses, pharmacists and physicians survey questionnaires: From the 200 staff sent questionnaires a total of 149 respondents comprising nurses (52.3%), physicians (32.2%) and pharmacists (16.1%) returned the questionnaires a total response rate of 74.5%. All responses were analyzed and compared item-by-item to see if there were any significant differences between the three groups for each questionnaire item. All three groups were most in agreement about their perception of hospital administration as making patient safety a top priority with regard to communicating with staff and taking action when medication errors were reported (all means 3.0 and p > 0.05). Pharmacists were most assured of administration support when an error was reported whereas nurses were least likely to see the administration as being supportive ( p < 0.001), and were more afraid of the negative consequences associated with reporting of medication errors (p = 0.026). Although nurses were generally less likely to perceive themselves as being able to communicate freely regarding reporting of errors compared to pharmacists there was no significant difference between the two groups. Both however were significantly different from physicians (p< 0.001). Physicians had the most favorable response to perceiving new technology as helping to create a safer environment for patients and to the full utilization of such technologies within the institution in order to help prevent medical errors. Scenario response - Responses to two scenarios outlining possible consequences, should a staff member commit a medication error, tended to be very similar among the three groups and followed the same general trend in which the later the error was discovered and the more grievous the patient harm, the more severe would be the consequences to the staff member. Interestingly, physicians saw themselves as less likely to suffer consequences and nurses saw themselves as more likely to suffer consequences should they have committed a medication error. All three groups were more likely to see themselves as facing dismissal from their job if the patient were to die. RESULTS OF ALL THREE OBSERVATIONS: Result of Nursing observations: For 1124 doses studied, 194 resulted in some form of error. The error rate was 17.2% and the accuracy was 82.8%. The commonest errors in a descending order were: wrong time, wrong drug, omission, wrong strength/ dose, wrong route, wrong instruction and wrong technique. No wrong drug form was actually administered in the observational period. These were the total number of errors observed for the entire month period of the study. IV Result of Pharmacist observations: A total of 2472 doses were observed during the one month period. Observations were done for 3 hours per day each day that the study was carried out. The study showed that there were 118 errors detected which were in the following categories respectively: 52 no instructions, 28 wrong drug/unordered, 21 wrong strength/dose, ignored/omission 13, shortage of medication 3 and expired date 1. Result of Prescribers in Chart review for drug-drug interactions: The analysis of the drug-drug interactions showed that out of a total of 1000 prescriptions, 124 had drug-drug interactions. None were found to fall into the highest severity rating i.e. 4 (contraindicated). Only twenty-one interactions were rated 3 (major), 87 interactions were rated moderate and 15 interactions were rated minor according the modified Micromedex scale. Patient education: All health care such as physician, pharmacist, and nurses have a responsibility to educate patient about their medication use and their health conditions to protecting them from any error can occur by wrong using drugs. Conclusion This study has contributed to the field of medication errors by providing data for a Middle Eastern country for the very first time. The views and opinions of the nurses, pharmacists and physicians should be considered to enhance the systems to minimize any errors in the future.

Patientsäkerhet är ett viktigt ämne inom hälso- och sjukvården. Läkemedelshantering är en del av omvårdnadsarbetet som kan äventyra patientsäkerheten. En fjärdedel av de vårdskador som inträffar beror på fel i läkemedelshanteringen. Studiens syfte var att beskriva händelser och/eller omständigheter i omvårdnadsarbetet som kan riskera patientsäkerheten vid läkemedelshantering. Metoden är litteraturstudie och omfattar 17 artiklar i resultatet som visar att det är flera händelser/omständigheter som bidrar till och ibland samverkar till att medicineringsfel inträffar. Händelser/omständigheter som kunde utgöra en risk för medicineringsfel var: hög arbetsbelastning, övertid, långa arbetspass, kunskapsbrist, låg erfarenhet, kommunikationsbrister, avbrott/distraktion, otydliga ordinationer, bristande kontrollåtgärder av sjuksköterskan samt relation och ansvarsfördelning mellan personalen. Fortsatt forskning bör fokusera på metoder, hjälpmedel och system som kan förhindra att medicineringsfel inträffar och på det sättet spara både pengar och mänskligt lidande. Det är även väsentligt att utvärdera effekterna av den nya patientsäkerhetslagen. / Patient safety is important for health care. Handling of pharmaceutical products is a task within nursing that may jeopardize patient safety. One fourth of the all documented health damages that occur are due to medication errors. The purpose of this study was to describe events and/or circumstances in nursing that can jeopardize patient safety when it comes to handling of pharmaceutical products. The study design is literature study which covers 17 articles in it’s result that shows it’s several events/circumstances that contributes and sometimes work together to make medication errors occur. Events or circumstances that could pose a threat to medication errors was: high workload, overtime, long shifts, lack of knowledge, lack of experience, communication gaps, interruption/distraction, unclear prescription, lack of control measures from the nurse and the relation and responsibilities among staff. Further studies should focus upon methods, tools and systems that can prevent medication errors from occuring and thus save both money and human suffering. It’s also essential to evaluate the effects of the new law concerning patient safety.

Master of Pharmacy / Medicines are a fundamental healthcare intervention, but the benefits they provide depend entirely on the way in which they are used. This begins with prescribing, a complex task with substantial risks. Systematic evaluation of biomedical factors may be viewed as an essential component of this task, but prescribers also integrate an array of individual, social, cultural, environmental and commercial factors into their prescribing decisions. Furthermore, social and cultural characteristics of the prescriber’s workplace may influence how well prescribing decisions are carried out. Whilst numerous research efforts have helped to construct an in-depth understanding of non-biomedical influences on GP’s prescribing patterns, the characteristics of corresponding sorts of influences in teaching hospitals have not been well determined. In hospitals, supervised medical trainees, registrars and consultants prescribe within the framework of medicines management systems involving nurses, pharmacists and patients. Currently, little is known about whether each of these groups has distinct beliefs, attitudes and values that may affect either prescribing behaviour or how prescribing skills of medical trainees are acquired. The aim of this study was to explore the social and cultural dynamics of prescribing and prescriber training in teaching hospitals. To do this, established qualitative methods were employed. Junior doctors, registrars, consultants, nurses, and pharmacists from two metropolitan teaching hospitals were sampled purposively and invited to participate in semi-structured interviews. A brief questionnaire was used to collect demographic and contextual information. In the interviews, participants were asked about their attitudes towards prescribing, their perceptions of roles and responsibilities, how they communicated prescribing decisions, their perceptions of influences on prescribing, and their perceptions of factors contributing to prescribing errors. Participants were also asked for their opinions on various aspects of new prescriber training. Sampling proceeded until redundancy of themes was established. A pilot study was conducted with one participant from each professional group to optimise the interview schedule, and then using this tool, a further 38 participants were interviewed. In total, eight consultants, eight registrars, nine junior doctors, eleven pharmacists, and seven nurses participated. Using reiterative content analysis of a third of all transcripts, a coding scheme was developed, which was used to label and categorise the remaining transcripts. Categories were further developed and refined. The resultant core themes were cross indexed against the five different health professional types using thematic charts to explore patterns. The main lines of enquiry for this research were mapped, the properties of these categories and interrelationships explored in detail, and a model of the prescribing process was developed. Prescribing at the teaching hospitals was a complex process consisting of multiple steps undertaken by several different health professionals of varying levels of experience from three different health care disciplines. Because of the intricate separation of responsibilities, the operation of the process was highly reliant on the behaviours of each player and their relationships with each other. Key prescribing decisions associated with patient admissions were made, almost exclusively, by medical teams. Prescribing was therefore chiefly characterised by factors influencing the behaviours of the doctors. Their behaviours were influenced by factors relating to their individual characteristics (eg, knowledge, skills, experience); but also by a web of socio-cultural determinants inherent to the environment in which they worked. These factors were related to: the organisational structure of the prescribing process; the knowledge characteristics of the doctors; the communication patterns they used; the underlying assumptions they made about prescribing; and the work environment.

Master of Pharmacy / Medicines are a fundamental healthcare intervention, but the benefits they provide depend entirely on the way in which they are used. This begins with prescribing, a complex task with substantial risks. Systematic evaluation of biomedical factors may be viewed as an essential component of this task, but prescribers also integrate an array of individual, social, cultural, environmental and commercial factors into their prescribing decisions. Furthermore, social and cultural characteristics of the prescriber’s workplace may influence how well prescribing decisions are carried out. Whilst numerous research efforts have helped to construct an in-depth understanding of non-biomedical influences on GP’s prescribing patterns, the characteristics of corresponding sorts of influences in teaching hospitals have not been well determined. In hospitals, supervised medical trainees, registrars and consultants prescribe within the framework of medicines management systems involving nurses, pharmacists and patients. Currently, little is known about whether each of these groups has distinct beliefs, attitudes and values that may affect either prescribing behaviour or how prescribing skills of medical trainees are acquired. The aim of this study was to explore the social and cultural dynamics of prescribing and prescriber training in teaching hospitals. To do this, established qualitative methods were employed. Junior doctors, registrars, consultants, nurses, and pharmacists from two metropolitan teaching hospitals were sampled purposively and invited to participate in semi-structured interviews. A brief questionnaire was used to collect demographic and contextual information. In the interviews, participants were asked about their attitudes towards prescribing, their perceptions of roles and responsibilities, how they communicated prescribing decisions, their perceptions of influences on prescribing, and their perceptions of factors contributing to prescribing errors. Participants were also asked for their opinions on various aspects of new prescriber training. Sampling proceeded until redundancy of themes was established. A pilot study was conducted with one participant from each professional group to optimise the interview schedule, and then using this tool, a further 38 participants were interviewed. In total, eight consultants, eight registrars, nine junior doctors, eleven pharmacists, and seven nurses participated. Using reiterative content analysis of a third of all transcripts, a coding scheme was developed, which was used to label and categorise the remaining transcripts. Categories were further developed and refined. The resultant core themes were cross indexed against the five different health professional types using thematic charts to explore patterns. The main lines of enquiry for this research were mapped, the properties of these categories and interrelationships explored in detail, and a model of the prescribing process was developed. Prescribing at the teaching hospitals was a complex process consisting of multiple steps undertaken by several different health professionals of varying levels of experience from three different health care disciplines. Because of the intricate separation of responsibilities, the operation of the process was highly reliant on the behaviours of each player and their relationships with each other. Key prescribing decisions associated with patient admissions were made, almost exclusively, by medical teams. Prescribing was therefore chiefly characterised by factors influencing the behaviours of the doctors. Their behaviours were influenced by factors relating to their individual characteristics (eg, knowledge, skills, experience); but also by a web of socio-cultural determinants inherent to the environment in which they worked. These factors were related to: the organisational structure of the prescribing process; the knowledge characteristics of the doctors; the communication patterns they used; the underlying assumptions they made about prescribing; and the work environment.

This research has been undertaken to evaluate factors with an association with the practice of self-medication amongst respondents living within three cities within Kurdistan. The research was designed to be a cross-sectional one by arranging for data collection through the direct interviewing of respondents via the use of a questionnaire that had been prepared previously. In total, the investigation involved 627 pharmacist participants, 647 general participants, and 28 interviewees from various age groups. An explanatory design is a mixed methods approach with two phases, with quantitative data collection in the first phase, and qualitative data collection in a second; data collection was conducted using a non-probability convenience sampling technique. The primary reason for self-medication practice was that participants with previous experience of attending to the same disease. The information source regarding self-medicated drugs were previous prescriptions, community pharmacies and friends. The most common indication for self-medication was the common cold or fever/headache/infection, the drugs used to treat these conditions being most commonly antibiotics, then painkillers and preparations for coughs. From the general public, a sample of 647 participants was taken that consisted of 38.4% females and 61.6% males, with participant ages ranging from 18-70 years. Within the study, 12.4% of the cohort had a degree level of education from a university. Moreover, 243 participants had the belief that it was an acceptable practice to purchase antibiotics without a prescription. Self-medication was practiced by 14/28 of the interviewees, and 28/28 (100%) of the interviewees held the belief that the pharmacy always has someone with knowledge of medicines, and who can advise and provide medication. There were 627 pharmacist participants, of which 28.1% were female and 71.9% were male, and 57.2% of them holding a Diploma in Health Institution, and 39.2% of them having a Bachelors Pharmacy degree. 20.7% of participants disagreed with keeping records for the dispensing of drugs, and approximately 20% of participants had little or no ideas regarding the characteristics of pharmacy practice that are considered professional. It was discovered that, if asked by the customer, advice was provided by 82.5% of community pharmacists. The sale of antibiotics was the most common, followed by pain-killers. A 95.5% proportion of pharmacists sold all of the medicines as OTC medicine without prescription. In conclusion, medicines are used by the people of Kurdistan in an inventive way, with suggestions provided by lay people and members of family or friends, which is acted upon without a qualified healthcare professional being consulted.

Current status: Medication errors within the context of safe and high quality health care are one of the most frequently discussed topics of recent years. They are part of the medical process, occur in different forms, affect patients of all ages and decrease the quality of provided health care. Prescriptions of medication in medical documentation have to be clearly legible and include all the required information. Nurses' role within the medicaiton process could be described as an ultimate "safeguard" that can prevent medication errors. Goals: The objectives of this research are to assess nurses' awareness of medication errors, ssess their ability to detect medication errors and find out the level of their knowledge regarding the correct procedures and measures following medication error detection.Methodology: The research was conducted using a qualitative research method. The in-depth interviews with nurses were carried out at a surgical critical care unit at the hospital in Jindrichuv Hradec. Following the collection of a sufficient amount of data the interviews were coded using the "pen and pencil" technique and then divided into categories.Research file: In order to achieve the highest level of objectivity the research was carried out with a group of nurses who work at the same unit of a hospital department. The research file included nurses who had varied levels of education and numbers of years of experience. Results: Surgical critical care unit nurses are not aware of the exact definition of medication errors. Only one nurse was able to discuss adverse events with related consequences. The remaining nurses' awareness of medication errors corresponded to the classification described in the relevant foreign and Czech academic literature. None of the interviewed nurses detected all the errors included in the simulated scenarios. Only one nurse, who is currently completing her qualification in this specialty, detected a wrongly prescribed antidote. The simulated scenarios also included a group of high-risk medications that is commonly used at the unit. The first medication included in this group was a 7.45% solution of potassium (KCl). A majority of the participants detected the high concentration of this high-risk medication. Another high-risk medication was a 10% concentrate of NaCl in an infusion; this error was also detected by the majority of the nurses. The third high-risk medication was a wrongly prescribed insulin (it lacked the detailed description of units, time and route). This prescription error was not detected by the majority of participating nurses. Another type of medication error included in the simulated scenarios was a group of medications used specifically in critical care. Only half of the participating nurses successfully detected the errors related to the prescription of this group of medications. The last type or medication errors focused on incomplete prescription by doctors. The most frequently detected error was a prescription of an opiate that lacked the route and the least frequently detected error was a wrong prescription of an infusion administration. It is evident that the nurses always inform the doctor when they detect a medication error. They are more willing to inform the ward sister in cases of incidents caused by somebody else. Only a small group of nurses are willing to inform the senior consultant, the head nurse or the hospital management. Only one nurse associated medication error reporting with an audit. However the majority of nurses agrees that it is important to report any medication errors.Conclusion: The analysed data suggested that the nurses were aware of medication errors. The analysis of the results of the simulated scenarios suggested that nurses' ability to detect medication errors in prescriptions was related to the number of years of experience and the level of education.

Abstract Background and Objective: An accurate medication history list is an integral part of the patient assessment at hospital admission. The objective of the study was to describe the frequency, type, and predictors of unintentional medication errors and to evaluate the quality of the clinical pharmacy services focusing on the acceptance of the recommendations made by the clinical pharmacist. Setting and methods: A descriptive study was conducted at two internal medicine wards at Baghdad Teaching Hospital in Iraq using Lund Integrated Medicines Management (LIMM)-based medication reconciliation. The study pharmacist conducted medication interviews for patients shortly after hospital admission to obtain the most accurate pre-admission medication history list. This list was compared with the medication list in the patient’s medical chart. Intended addition, withdrawal of a drug, or changes to the dose/ dosage form in the patient’s medical list was considered as medication discrepancies. However, medication discrepancies were considered as medication errors based on no identified clinical reason. Results: A total of 114 patients were included in this study. Over two-thirds of the study patients (73.7%) experienced 215 medication errors identified by a clinical pharmacist conducting medication reconciliation. Most errors were omission (87.9%). Cardiovascular agents followed by NSAID were commonly in error (53%) and (10.2%) respectively. In a logistic regression model, age (odds ratio (OR), 1.055: 95% confidence interval (CI) 1.010 - 1.102), female gender (OR, 3.468: 95% CI 1.232- 9.761) and number of medications at admission (OR, 0.810: 95% CI 0.681-0.963) were predictors for medication history errors at admission. Conclusions: Medication errors at the time of hospital admission are common and undetected.  A structured approach like the LIMM-based medication reconciliation at Iraqi hospital is needed to detect these errors.

Este estudo identificou, analisou e comparou os erros de medicação ocorridos nos processos de preparo e administração de medicamentos, em uma unidade de internação clínica de um hospital público de ensino, pertencente à Rede de Hospitais-Sentinela da Agência Nacional de Vigilância Sanitária (ANVISA) e localizado na cidade de Rio Branco-AC. Trata-se de um estudo observacional e transversal, realizado no período de julho a setembro de 2005. A amostra foi constituída de 1.129 doses de medicamentos. Esta investigação foi desenvolvida em duas fases: na primeira, os dados foram obtidos a partir da observação direta dos processos que compõem o sistema de medicação e de entrevistas com três profissionais: o responsável pelo setor de farmácia, o chefe do serviço médico e a supervisora do serviço de enfermagem. Na segunda fase, foram observados o preparo e a administração de 1.129 doses, e os erros de medicação foram identificados. Os resultados permitiram identificar 404 (35,8%) erros de medicação e um sistema de medicação com 56 ações. Verificou-se que 866 (76,7%) prescrições estavam manuscritas; 126 (11,2%) não continham o nome legível do medicamento; em 267 (23,6%) faltavam as doses; em 107 (9,5%) não constava a via; em 712 (63,1%) não havia a forma de apresentação; em 20 (1,8%) faltava a freqüência; e em 338 (29,9%) não constavam o tipo e volume do diluente para o preparo. No preparo de medicamentos, foi identificado que 976 (86,4%) doses estavam rotuladas incorretamente e 49 (4,3%) doses não possuíam rótulo. Em relação à administração, observou-se que apenas 31 (2,7%) doses foram administradas após conferência direta da prescrição; em 691 (61,2%) doses não ocorreu identificação do paciente e em 904 (80,1%) doses não houve orientação a respeito do medicamento. Constatouse que 179 (78,2%) doses infundidas não foram controladas, e 214 (18,9%) doses foram registradas imediatamente após a administração. Nas observações em que se desconhecia previamente a prescrição do medicamento, ocorreram os seguintes erros: 47 (4,2%) erros de dose, 2 (0,2%) erros de via, 130 (11,5%) erros de horário, 2 (0,2%) erros de pacientes, 11(1%) erros de medicamentos não autorizados e 71 (6,3%) erros de omissão. Nas observações em que se conhecia previamente a prescrição, identificaram-se 17 (1,5%) erros de dose, 85 (7,5%) erros de horário, 4 (0,4%) erros de medicamentos não autorizados e 35 (3,1%) erros de omissão. Nessa etapa, não ocorreram erros de via e de paciente. Propõe-se como medidas para a redução dos erros nessa instituição: formar um grupo multiprofissional com a finalidade de discutir e estabelecer estratégias que possam promover a segurança do paciente; elaborar protocolos de preparo e administração de medicamentos e promover a educação continuada e permanente para os profissionais. Sugere-se, ainda, que a instituição padronize a prescrição médica, normatizando os itens da prescrição dos medicamentos; desenvolvendo um sistema de distribuição de dose unitária e implementando a prescrição médica eletrônica. / This study identified, analyzed and compared the medication errors that occurred in the medication preparation and administration processes at a clinical hospitalization unit of a public teaching hospital, which is part of the Sentinel Hospital Network of the Brazilian National Health Surveillance Agency (ANVISA) and located in Rio Branco-AC, Brazil. We carried out an observational and cross-sectional study between July and September 2005. The sample consisted of 1,129 medication doses. This study was developed in two phases: in the first, data were obtained through direct observation of the medication system processes and interviews with three professionals: the pharmacy sector responsible, the medical service head and the nursing service supervisor. In the second phase, we observed the preparation and administration of 1,129 doses and identified medication errors. The results revealed 404 (35.8%) medication errors and a medication system that consisted of 56 actions. We found 866 (76.7%) handwritten prescriptions; 126 (11.2%) did not contain the readable name of the drug; doses were missing in 267 (23.6%); route in 107 (9.5%); form in 712 (63.1%); frequency in 20 (1.8%); and the diluent type and volume for preparation in 338 (29.9%). In medication preparation, we identified that 976 (86.4%) doses were labeled incorrectly and that 49 (4.3%) doses did not have a label. With respect to administration, only 31 (2.7%) doses were administered after direct verification of the prescription; in 691 (61.2%) doses, the patient was not identified and, in 904 doses (80.1%), no orientation was provided about the drug. We found that 179 (78.2%) infused doses were not controlled, and that 214 (18.9%) were registered immediately after their administration. In those observations when the medication prescription was previously unknown, the following errors occurred: 47 (4.2%) dose errors, 2 (0.2%) route errors, 130 (11.5%) time errors, 2 (0.2%) patient errors, 11(1%) unauthorized medication errors and 71 (6.3%) omission errors. In those cases when the medication prescription was previously known, we identified 17 (1.5%) dose errors, 85 (7.5%) time errors, 4 (0.4%) unauthorized medication errors and 35 (3.1%) omission errors. In this phase, no route and patient errors occurred. To reduce errors at this institution, we propose the following measures: constitute a multiprofessional group to discuss and establish strategies with a view to promoting patient safety; elaborate medication preparation and administration protocols and promote continuing and permanent professional education. We also suggest that the institution should standardize medication prescriptions by normalizing medication prescription items; developing a unit dose distribution system and implementing electronic medical prescriptions.

Os erros na medicação podem trazer sérias conseqüências aos pacientes, profissionais e às instituições de saúde; resultando de múltiplas causas, dentre elas: falhas profissionais e do sistema de medicação. O objetivo deste estudo foi identificar e analisar os processos do sistema de medicação, suas falhas e propor medidas de melhorias ao hospital. Realizou-se uma pesquisa quantitativa do tipo survey exploratório, na unidade de clínica médica e na farmácia de um hospital geral e universitário do estado de Goiás, após aprovação do Comitê de Ética em Pesquisa. A amostra constituiu-se de um farmacêutico na primeira etapa, 40 profissionais na terceira etapa, sendo 12 (30%) médicos-residentes; 20 (50%) profissionais de enfermagem e 8 (20%) profissionais da equipe de farmácia, além da utilização de 294 prontuários. Realizou-se a coleta dos dados em 2002, por meio de entrevista com o responsável pelo sistema de medicação da instituição; de observação não-participante de ambientes e de ações dos profissionais e acadêmicos; de entrevista com profissionais e de análise de prontuários. Consultaram-se todos os profissionais quanto à sua disposição para participar do estudo e, a seguir, solicitou-se a assinatura do termo de consentimento livre e esclarecido. Os dados obtidos possibilitaram caracterizar os seguintes processos: prescrição de medicamentos realizada manualmente com cópia carbonada, dispensação por dose individualizada, além de indicar que o profissional de enfermagem que administra o medicamento não é o mesmo que o prepara, na clínica médica. Os resultados das observações realizadas durante 21 dias, nos processos de prescrição, dispensação e administração de medicamentos, indicaram: o ambiente como principal problema no processo de prescrição (69%) e de dispensação de medicamentos (30,6%), por se tratar de local impróprio, com ruídos e interrupções freqüentes; falhas na segurança durante a técnica e preparo antecipado do medicamento no processo de preparação (46,8%). O estudo possibilitou, também, a construção de um gráfico do fluxo das 60 ações desenvolvidas desde a prescrição até o monitoramento. As entrevistas com os profissionais apontaram que os tipos de erros mais freqüentes estavam relacionados à prescrição médica (29%) e ao horário (20,6%); suas causas deviam-se a falhas individuais e falta de atenção (47,4%) e excesso de trabalho (14,5%). As falhas individuais dos profissionais foram apontadas também como principal falha do sistema de medicação (27%). A alteração nas atitudes individuais foi a sugestão mais indicada para evitar a ocorrência de erros, com 28,3% das respostas, e a orientação, a providência mais tomada (25%). A análise dos prontuários mostrou prescrições de medicamentos com 64,6% de legibilidade, tendo 62,2% tanto nomes de medicamentos comerciais quanto do princípio ativo; 95% estavam incompletas para algum item; 96% apresentavam abreviaturas e 30% rasuras. Encontraram-se anotações sobre medicamentos apenas em sete relatos de enfermagem e uma outra na evolução médica. A clínica não possui relatório de ocorrências sobre erros na medicação. As medidas propostas para melhorar o sistema e, conseqüentemente, prevenir erros na medicação foram: prescrição eletrônica, dose unitária, relatórios sobre erros, cultura não-punitiva, segurança do paciente e, enfim, simplificação do sistema. / The medication errors can bring serious consequences to patients, professionals and healthcare institutions, they have multiple causes, amongst them failures related to the professionals and related to the medication system. This study’s objective was to identify and to analyse the medication system process, its failures in order to propose improvement actions to the hospital. This exploratory descriptive study took place in the medical clinical unit and in the pharmacy of a general and university hospital of the state of Goiás, after approval of the Committee of the Ethics of the correspondent hospital. The sample included: (first phase) a pharmacist, (third phase) 40 professionals divided into 12 resident physicians (30%), 20 nursing professionals (50%), 8 pharmacy team members (20%); 294 patient charts were also used. The data was collected in 2002 and consisted of an interview with the professional in charge of the medication system, and non-participant observation of the environment and actions of the professionals and the academic people, interviews with the professionals and patient chart analysis. The professionals were asked to sign on the “Free Will Participation Agreement”. It was possible based on the data collected to describe the following processes: handwriting medication prescription using carbon paper, individually dose dispensing and that the nursing professional who administrates the drug is not the same that prepares it in the clinical unit. The results based on the 21 days of observation of the drug prescription, dispensing and administration processes were: the environment is the main problem in the prescription (69%) and dispensing (30,6%) processes, it is a noisy place and interruptions frequently occur; safety failures during the technique and in-advance drug preparation appeared in the top (46,8%) in the preparation process. The data collection also allowed to build a chart of the 60 steps from drug prescription to monitoring. The results from the interviews showed that the most frequent errors were related to both prescription (29%) and schedule (20,6%) and their main cause were individual failures and lack of attention (47,7%) and work overload (14,5%). The individual failures were also listed as the main failure in the medication system (27%). In order to avoid errors 28,3% of the answers suggested to change the individuals’ behavior, and orientation as the administrative action more frequently taken (25%). The patient chart analysis found out the following drug prescriptions characteristics: 64,6% readable, 62,2% using drug brand names as well as the active principle name, 95% incomplete for missing information, 96% using abbreviations and 30% with erasures. Concerning to drug notifications, the analysis also found out 7 nursing reports, one from the physician and that there is no error report in the clinical unit. The suggested improvements to avoid errors and enhance the system are: computerized physician order electronic entry, unit dose, errors reports, non-punitive approach, patient safety, and at last to make the system as simple and lean as possible.

Eventos adversos aos medicamentos e erros de medicação são muito comuns na prática assistencial e podem ocorrer em qualquer etapa do processo da terapia medicamentosa, contribuindo com a ocorrência de iatrogenias nos pacientes devido ao uso incorreto dos medicamentos ou a sua omissão. Neste contexto, encontramse as prescrições médicas que têm papel ímpar na prevenção do erro, uma vez que prescrições ambíguas, ilegíveis ou incompletas podem contribuir com a ocorrência destes eventos. Esta investigação teve o propósito de analisar a redação de prescrições médicas em unidades de clínica médica de cinco hospitais Brasileiros, comparar os dados obtidos entre os hospitais e propor recomendações para a prevenção de futuros erros de medicação. Tratou-se de um estudo descritivo que utilizou de dados secundários obtidos de uma pesquisa multicêntrica realizada em 2005. A população foi composta por 1.425 medicamentos administrados em discordância com a prescrição. Deste total, a administração de medicamentos em horário diferente do prescrito foi o mais freqüente nos cinco hospitais investigados, correspondendo a 76,0%. A análise da redação da prescrição revelou que 93,6% continham siglas e/ou abreviaturas, 10,7% não apresentavam dados do paciente, 4,3% omitiram informações sobre o medicamento e 4,2% apresentavam alterações e/ou suspensão do medicamento. Com a implantação do sistema computadorizado de prescrições, associada à prática da educação continuada e permanente dos profissionais envolvidos no sistema de medicação será possível minimizar os danos causados aos pacientes hospitalizados decorrentes da administração de medicamentos e, consequentemente, melhorar a qualidade do cuidado prestado. / Adverse events related to medicines and medication errors are very common in the health care assistance and can occur at any stage of the medication process, contributing with the occurrence of iatrogenys in the patients due to the incorrect use of medicines or its omission. In this context, we find medical orders which has uneven role in the prevention of medication error, once ambiguous, unreadable or incomplete medication order may contribute with the occurrence of these events. This study had the intention to analyze the writing of medical orders at internal medicine units of five Brazilian hospitals, to compare data between these hospitals and to consider recommendations for the prevention of future medication errors. This descriptive study used secondary data from a multicentric research occurred in 2005. The population was composed of 1.425 medications given in discordance with the medical order. From this total, the medication administration at different schedule administration time was the most frequent error found at the five hospitals investigated, corresponding to 76.0%. The analysis of the writing of the medical order disclosed that 93.6% contained acronyms and/or abbreviations, 10.7% did not present any information about the patient, 4.3% had omitted information about the medicine and 4.2% presented alterations and/or suspension of the medicine. With the implantation of the computerized prescription order entry system, associate to the practice of continued and permanent education of the involved professionals in the system of medication it will be possible to minimize the damage caused to the patients in the hospital deriving from the administration of medicine and, therefore, improve the quality of the care given.