Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-01-26T15:26:50Z No. of bitstreams: 1 EricoDeLimaVale_DISSERT.pdf: 1912656 bytes, checksum: d9fad5e12618fef8d0405ac649e8f9df (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-01-26T15:45:09Z (GMT) No. of bitstreams: 1 EricoDeLimaVale_DISSERT.pdf: 1912656 bytes, checksum: d9fad5e12618fef8d0405ac649e8f9df (MD5) / Made available in DSpace on 2017-01-26T15:45:09Z (GMT). No. of bitstreams: 1 EricoDeLimaVale_DISSERT.pdf: 1912656 bytes, checksum: d9fad5e12618fef8d0405ac649e8f9df (MD5) Previous issue date: 2016-08-01 / Objetivos: Realizar um ciclo de melhoria da qualidade em uma unidade de terapia intensiva materna (UTIM) e avaliar seu impacto na assist?ncia multiprofissional ?s pacientes com doen?as hipertensivas gestacionais (DHG). M?todos: Foi realizado um ciclo de melhoria de maio a julho de 2015; os per?odos pr? e p?s interven??o foram de janeiro a abril e de agosto a outubro do mesmo ano, respectivamente. Os crit?rios definidos para avalia??o foram: (1) solicita??o de exames laboratoriais quando da admiss?o na UTIM; (2) solicita??o de ultrassom obst?trico quando da admiss?o na UTIM; (3) controle de picos press?rico com uso de Hidralazina intravenosa;(4) uso de anti-hipertensivos orais para controle press?rico; (5) uso de Inibidores da Enzima Conversora da Angiotensina (IECA) ou Bloqueadores do Receptor da Angiotensina (BRA); (6) restri??o h?drica intravenosa; (7) indica??o do corticoide Betametasona em pacientes com idade gestacional menor que 35 semanas; (8) uso do sulfato de magn?sio (MgSO4) e (9) manuten??o do MgSO4 no p?s parto. Todas as mulheres admitidas na UTIM com diagn?stico de DHG nos per?odos pr? e p?s interven??o foram eleg?veis para o estudo. Foram analisadas as admiss?es antes (50) e ap?s (50) a realiza??o do ciclo de melhoria da qualidade. O desfecho avaliado foi a taxa de adequa??o total e individual das recomenda??es baseadas em evid?ncias nas pacientes com DHG. Em cada avalia??o foram calculados os intervalos de confian?a de 95% para as estimativas de conformidade, suas diferen?as absoluta e relativa e o valor Z (uma cauda), sendo considerado significativo valor de p <0,05. Resultados: Houve aumento da taxa total de adequa??o dos crit?rios (p1=88+3%, p2=92+1%; p=0,018) e solicita??o de ultrassom fetal (p1=72+10%, p2=88+4%; p=0,023), e redu??o no uso de anti-hipertensivos orais (p1=100%, p2=94+3%; p=0,039), n?o houve altera??es significativas nos demais crit?rios. Conclus?o: A realiza??o de um ciclo de melhoria est? associado a um aumento na taxa de ades?o as recomenda??es baseadas em evid?ncia para o tratamento de pacientes com DHG. / Objectives: To conduct a quality improvement cycle in a maternal intensive care unit (MICU) and assess their impact on multidisciplinary care for patients with gestational hypertensive disease (GHD). Methods: An improvement cycle was conducted from May to July 2015; pre and post intervention were from January to April and from August to October of that year, respectively. The criteria for evaluation were: (1) request for laboratory tests at admission in MICU; (2) obstetrical ultrasound request when admission to the MICU; (3) control of pressure peaks with the use of intravenous hydralazine; (4) use of oral antihypertensive drugs for blood pressure control; (5) use of inhibitors of Angiotensin Converting Enzyme (ACE) inhibitors or Angiotensin Receptor Blockers (ARBs); (6) intravenous fluid restriction; (7) indication of betamethasone steroids in patients with gestational age less than 35 weeks; (8) use of magnesium sulphate (MgSO4) and (9) MgSO4 maintenance postpartum. All women admitted in MICU diagnosed with GHD pre and post intervention were eligible for the study. The implementation of the recommendations was investigated before (n = 50) and after (n = 50) the implementation of the quality improvement cycle. The primary outcome was the rate of overall and individual adherence to evidence-based recommendations in patients with GHD. In each evaluation were calculated 95% confidence intervals for the estimates of compliance, their absolute and relative differences and the Z value (one tail), being considered significant an p <0.05. Results: There was increase in total adherence ratio (p1 = 88 + 3%, p2 = 92 + 1%; p = 0.018) and individual fetal ultrasound request (p1 = 72 + 10%, p2 = 88 + 4%; p = 0.023), and a reduced use of oral anti-hypertensives (p1 = 100%, p2 = 94 + 3%; p = 0.039), there were no significant changes in other criteria. Conclusion: The completion of a quality improvement cycle was associated with an increase in the adhesion rate of the evidence-based recommendations for the treatment of patients with GHD.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-02-02T15:04:34Z No. of bitstreams: 1 LuziaClaraCunhaDeMenezes_DISSERT.pdf: 770807 bytes, checksum: cdc5621b5f67cf18ce66f51850c560bc (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-02-09T19:57:41Z (GMT) No. of bitstreams: 1 LuziaClaraCunhaDeMenezes_DISSERT.pdf: 770807 bytes, checksum: cdc5621b5f67cf18ce66f51850c560bc (MD5) / Made available in DSpace on 2017-02-09T19:57:41Z (GMT). No. of bitstreams: 1 LuziaClaraCunhaDeMenezes_DISSERT.pdf: 770807 bytes, checksum: cdc5621b5f67cf18ce66f51850c560bc (MD5) Previous issue date: 2016-08-19 / Os servi?os de sa?de t?m cada vez mais incorporado novas tecnologias e t?cnicas acompanhadas de riscos adicionais ? seguran?a dos pacientes. Pacientes internados em Unidade de Terapia Intensiva (UTI) est?o mais suscept?veis a erros. Dentre as ferramentas utilizadas no mundo para reduzir a ocorr?ncia ou recorr?ncia desses erros na assist?ncia ? sa?de, destacam-se a An?lise de Causa-Raiz (ACR) e An?lise Modal de Falhas e Efeitos (AMFE). O objetivo do estudo foi identificar e analisar a aplica??o das ferramentas ACR e AMFE para a melhoria da qualidade da assist?ncia em UTI. Revis?o sistem?tica de literatura baseada no protocolo PRISMA. Foram utilizadas as seguintes bases de dados: Scopus, PubMed, SciELO, LILACS, Web of Science, Science Direct, Cochrane, WHOLIS, PAHO e EMBASE. A an?lise qualitativa dos artigos foi realizada atrav?s da aplica??o de uma vers?o adaptada e resumida do guia SQUIRE 2.0. Inclu?dos artigos originais indexados nas bases de dados e que cumprissem 60% dos crit?rios do SQUIRE 2.0 adaptado. Foram recuperados 1674 documentos nas buscas e, ap?s as an?lises pertinentes, foram inclu?dos 18 artigos na revis?o. Desses, 16 foram publicados entre 2010 e 2016, 10 foram desenvolvidos nos Estados Unidos, 11 foram realizados em UTI Pedi?trica ou Neonatal, 16 utilizaram AMFE em diversas tem?ticas, 13 realizaram interven??es para melhoria da qualidade e desses, 12 utilizaram indicadores para medi??o da melhoria relatando resultados positivos. Esses dados sugerem preocupa??o com o planejamento da qualidade, sendo importante ressaltar a utiliza??o de indicadores para medir a melhoria da qualidade na maioria dos estudos selecionados. Essa revis?o refor?a a import?ncia do uso dessas ferramentas para a melhoria da qualidade da assist?ncia nas UTI, permeando as institui??es de sa?de de comportamentos que garantam mais seguran?a, contribuindo para o desenvolvimento de uma cultura de seguran?a organizacional. / Health services have increasingly incorporated new technologies and techniques accompanied by additional risk to patient safety. Patients in Intensive Care Unit (ICU) are more susceptible to errors. Among the tools used in the world to prevent the occurrence or recurrence of these errors in health care, we highlight the Root Cause Analysis (RCA) and Failure Modes and Effects Analysis (FMEA). The objective of this study was to identify and analyze the application of RCA and FMEA tools for improving the quality of care in ICU. Systematic review of literature based on the PRISMA. We used the following data bases: Scopus, PubMed, SciELO, LILACS, Web of Science, Science Direct, Cochrane, WHOLIS, PAHO and EMBASE. The qualitative analysis of the articles was conducted by applying an adapted and abridged version of SQUIRE 2.0 guide. 1674 documents were recovered in searches and, after the relevant tests, 18 were included in the review articles. Of these, 16 were published between 2010 and 2016, 10 were developed in the United States, 11 were conducted in the Pediatric ICU or Neonatal, 16 used FMEA on different topics, 13 performed interventions for improvement of quality and of those, 12 have used indicators for measuring the improvement in reporting positive results. These data suggest concern with quality planning, it is important to highlight the use of indicators to measure the improvement of quality in most of the selected studies. This review underscores the importance of using these tools to improve the quality of care in the ICU, permeating the health institutions of behaviors that ensure more safety, contributing to the development of an organizational safety culture.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-03-21T18:34:54Z No. of bitstreams: 1 MabelMendesCavalcanti_DISSERT.pdf: 1030603 bytes, checksum: d08cb93653c93a011d5a53d2d6f4c87f (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-03-27T23:35:53Z (GMT) No. of bitstreams: 1 MabelMendesCavalcanti_DISSERT.pdf: 1030603 bytes, checksum: d08cb93653c93a011d5a53d2d6f4c87f (MD5) / Made available in DSpace on 2017-03-27T23:35:54Z (GMT). No. of bitstreams: 1 MabelMendesCavalcanti_DISSERT.pdf: 1030603 bytes, checksum: d08cb93653c93a011d5a53d2d6f4c87f (MD5) Previous issue date: 2016-10-14 / Introdu??o: Os servi?os de sa?de t?m crescente uso de tecnologias e processos complexos que oferecem riscos elevados durante o cuidado ao paciente. M?todos prospectivos de identifica??o e redu??o de riscos podem ser ?teis, por?m h? pouca descri??o de sua aplica??o no contexto brasileiro. Objetivo: esse estudo objetiva analisar, identificar e priorizar as falhas no processo de dispensa??o de medicamentos em um hospital universit?rio, com propostas de a??es para redu??o do risco das falhas priorizadas. M?todo: trata-se de um estudo descritivo, desenvolvido em um hospital universit?rio, com uso do HFMEA segundo a Veteran Affairs, aplicado por equipe multiprofissional, durante 3 meses. Foram identificados e analisados modos de falha no processo de dispensa??o de medicamentos, segundo crit?rios de gravidade, frequ?ncia, criticidade, chance de detec??o e exist?ncia de medidas preventivas. Com as falhas priorit?rias, foi realizada an?lise das causas e plano de gerenciamento e monitoramento dos riscos. Resultados: destacaram-se 21 modos de falha na dispensa??o de medicamentos. Os subprocessos com maiores falhas foram, em ordem decrescente, ?separa??o do medicamento? (06), ?prepara??o da dose? (06), ?avalia??o da prescri??o? (03), ?unitariza??o dos medicamentos? e ?entrega da dose? (02), ?libera??o do medicamento? (01) e ?confer?ncia da dose? (01). Tr?s dessas falhas foram selecionadas para an?lise e interven??o: identifica??o errada do medicamento (na unitariza??o), medicamento separado diferente do prescrito (na separa??o do medicamento) e atraso na entrega de dose (na entrega da dose). Ap?s a an?lise das causas, estabeleceu-se algumas interven??es necess?rias, a saber: auditoria para avaliar a ades?o ao protocolo do subprocesso ?unitariza??o? e divulga??o dos resultados, constru??o de um protocolo operacional padr?o para o armazenamento de medicamentos com som e grafia semelhantes e implanta??o do protocolo de seguran?a na prescri??o, uso e administra??o de medicamentos. Conclus?o: o estudo permitiu analisar o processo de dispensa??o de medicamentos, onde identificou-se as poss?veis falhas que pudessem aumentar o risco de eventos adversos nessa etapa do processo de medica??o. O HFMEA foi ?til para auxiliar o grupo multidisciplinar a entender melhor as fragilidades no processo de trabalho e favoreceu a percep??o de falhas espec?ficas do contexto do hospital participante, al?m de auxiliar na prioriza??o de interven??es corretivas e possibilidade de melhorias para seguran?a na dispensa??o de medicamentos. / Introduction: health services have increasing use of complex technologies and processes that provide high risks for patient care. Prospective methods of identification and risk reduction can be helpful, but there is little description of your application in the Brazilian context. Objective: this study aims to analyze, identify and prioritize failures in the drug dispensing process in a university hospital, with proposals for actions to reduce the risk of prioritized failures. Method: it is a descriptive study, developed in a university hospital, using the HFMEA according to Veterans Affairs, implemented by multidisciplinary team for 3 months. They were identified and analyzed failure modes in the drug dispensing process, according to their severity and frequency, from the priority analysis of causes, management and monitoring of risks. Results: the highlights 21 ways to fail in dispensing drugs. Subprocesses with major failures were, in descending order, separation of the product (06), prepare the dose (06), evaluation of the prescription (03), unitarization of drugs and delivery of the dose (02), the drug release (01) and conference dose (01). Three of these failures were selected for analysis and intervention: "misidentification of medicine" (unitarization), "different separate drug's prescribed" (separation of the product) and "delay in dose delivery" (the drug release). After the analysis of the causes, we established some necessary interventions, namely: audit to assess adherence to the sub-process protocol "unitarization" and dissemination of results, construction of a standard operating protocol for storing medicines with sound and the like spelling and implementation of the security protocol in prescription, use and administration of medications. Conclusion: the study allowed us to analyze the process of dispensing drugs, where we identified the possible failures that could increase the risk of adverse events in this stage of the medication process. HFMEA was useful to assist multidisciplinary group better understand the weaknesses in the work process, which can help to faults, and assists in prioritizing corrective interventions and possible improvements to safety in dispensing drugs.

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-07-04T15:02:40Z No. of bitstreams: 1 LucianoLuizDaSilvaJunior_DISSERT.pdf: 2133997 bytes, checksum: 0cac4f4fc4430aae598b37a984f70b9b (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-07-14T15:04:44Z (GMT) No. of bitstreams: 1 LucianoLuizDaSilvaJunior_DISSERT.pdf: 2133997 bytes, checksum: 0cac4f4fc4430aae598b37a984f70b9b (MD5) / Made available in DSpace on 2017-07-14T15:04:44Z (GMT). No. of bitstreams: 1 LucianoLuizDaSilvaJunior_DISSERT.pdf: 2133997 bytes, checksum: 0cac4f4fc4430aae598b37a984f70b9b (MD5) Previous issue date: 2016-08-23 / A aten??o ao c?ncer ? uma prioridade da sa?de p?blica mundial, por?m, ainda nos dias de hoje, persistem falhas simples capazes de repercutir em aumento da morbimortalidade dos pacientes. O registro inadequado de informa??es cl?nicas leva ? repeti??o de exames, erros de diagn?stico e tratamento, perda de tempo e custos desnecess?rios.Este trabalhoobjetivou verificara efic?cia de um ciclo de melhoria nos registros m?dicosde um hospital de refer?ncia em oncologia, avaliando o n?vel de qualidade no preenchimento dos prontu?rios, identificando os crit?rios de qualidade que mais apresentaram erros e analisando a efetividade dos ciclos de avalia??o cont?nua na melhoria da qualidade do preenchimento dos prontu?rios. Para tanto, dez crit?rios de qualidade em prontu?rios de pacientescom diagn?stico de c?ncer de pr?stata foram examinados. A partir de amostragens aleat?rias simples, procedeu-se ?aprecia??o desses crit?rios, antes e depois das interven??es propostas. Tais a??es envolveram administradores e m?dicos, sendo direcionadas especialmente ? conscientiza??o m?dica e reestrutura??o do prontu?rio por interm?dio de um checklist.No intuito de identificar o n?vel de qualidade, selecionaram-se a estimativa pontual e o intervalo de confian?a (95%) do cumprimento dos crit?rios. Para comprovar a efetividade das interven??es, foram calculadas as melhorias absoluta e relativa entre a primeira e ?ltima avalia??o, assim como a sua significa??o estat?stica (p<0,05). Na medi??o inicial, verificaram-se ?ndices de cumprimento de 66,4% na m?dia e um total de 411 de n?o conformidades, significando 34,2% de n?o cumprimento. Destacaram-se, como maior e menor percentual de n?o cumprimento, os crit?rios patologias associadas, 81,9%, e estadiamento?TNM, 9,5%. J?, na segunda medi??o, notou-se uma redu??o do n?mero total de defeitos em 49,4%. Em suma, a avalia??o baseada em crit?rios possibilitou a identifica??o de defeitos de qualidade no preenchimento dos prontu?rios levantados, o que, por sua vez, direcionou as interven??es propostas. Considerando as melhorias alcan?adas, percebe-se que o ciclo de melhoria foi efetivo na diminui??o dos defeitos observados, al?m de auxiliar expressivamente no progresso da qualidade do registro m?dico. / In spite of the fact that attention to cancer is a global public health priority, simple failures capable of resonating in the increase of morbidity and mortality of patients still persist nowadays. The Inadequate recording of clinical information leads to the repetition of tests, diagnostic and treatment errors, waste of time and unnecessary costs. This study aims at verifying the effectiveness of a cycle of improvement of medical records in a referral hospital in oncology, by evaluating the level of quality in filling in the patient?s medical records, identifying the quality criteria that presents the greater amount of errors and analyzing the effectiveness of the continuous evaluation cycles that contributed to the improvement of the quality of filling in the patient?s medical records. To this end, ten quality criteria in medical records of patients diagnosed with prostate cancer were examined. From simple random sampling, these criteria were assessed before and after the proposed interventions. Such actions involved administrators and doctors, being directed especially at medical awareness and restructuring of the patient?s medical records via a checklist. In order to identify the level of quality, the point-in-time estimate and confidence interval (95%) of the criteria compliance were selected. In order to prove the effectiveness of interventions, the absolute and relative improvements between the first and the final evaluation, as well as its statistical significance (p < 0.05) were calculated. Compliance rates of 66.4%, on average, and a total of 411 nonconformities, meaning 34.2% of non-compliance were verified in the initial measurement. The associated pathologies criteria, 81.9%, and staging - TNM, 9.5%, stood out as the highest and lowest percentage of non-compliance. In the second measurement, a reduction in the total number of defects of 49.4% was observed. In short, the criteria-based evaluation made possible the identification of quality defects in filling in the surveyed patient?s medical records, which, in turn, directed the proposed interventions. Considering the improvements achieved, it is reasonable to assume that the cycle of improvement was effective in decreasing the defects observed, besides contributing significantly to the progress of quality of the medical record.

Made available in DSpace on 2014-12-17T14:53:14Z (GMT). No. of bitstreams: 1 JaniereL.pdf: 2102402 bytes, checksum: 6a424cc3be8c6b5ca19e575a77e57039 (MD5) Previous issue date: 2006-05-19 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / This thesis presents a contribution to the study customer satisfaction models, analyzing the relationship between antecedent variables satisfaction and customer loyalty, through a survey with car s buyers in Natal. The theorical survey is focused in concepts of customer satisfaction and loyalty, and in the customer satisfaction index models. For the field survey, was applied a questionnaire, based on the Norwegian Customer Satisfaction Barometer (NCSB), considered by Johnson et al. (2001), with 106 customer of concessionary Fiat, the Pontanegra Autom?veis. The main results obtained by the multiple regression analysis reveal that, considering the Fiat, the satisfaction is influencing by reliance degree in the firm seriousness and by complaining handling, the loyalty is influenced by possibility of again pay the same value if was again buy a car, the satisfaction degree with a car, affective commitment relationship firm customer of the own car and the complaining handling. Considering the concessionaire, the satisfaction is influenced by reliance degree in the firm seriousness, ability degree of the concessionaire in delivery service and by complaining handling, already the loyalty is influencing by possibility of again pay the same value if was again buy a car, affective commitment relations to be concessionaire s customer, by calculate commitment in relation the economic loss of bought this car of this firm and complaining handling / Esta tese apresenta uma contribui??o ao estudo de modelos de satisfa??o do cliente, analisando as rela??es que envolvem as vari?veis antecedentes da satisfa??o e da fidelidade do cliente, atrav?s de uma pesquisa com compradores de carros em Natal. A pesquisa te?rica ? focada em conceitos de satisfa??o e fidelidade do cliente e nos modelos ?ndices de satisfa??o do cliente. Para a pesquisa de campo, foi aplicado um question?rio baseado no Modelo ?ndice de Satisfa??o Noruegu?s (NCSB), proposto por Johnson et al. (2001), a 106 clientes da Concession?ria Pontanegra Autom?veis. Os principais resultados obtidos pela regress?o m?ltipla, revelam que, levando em considera??o a Fiat, a satisfa??o ? influenciada pelo grau de confian?a na seriedade da empresa e pelo tratamento das reclama??es, a fidelidade ? influenciada pela possibilidade de pagar novamente o mesmo valor se fosse comprar de novo um carro, o grau de satisfa??o com o carro, o compromisso afetivo em ser cliente do fabricante do seu carro e o tratamento das reclama??es. Considerando a Concession?ria, a satisfa??o ? influenciada pelo grau de confian?a na seriedade da empresa, pelo grau de capacidade da concession?ria em prestar um servi?o e pelo tratamento das reclama??es. J? a fidelidade ? influenciada pela possibilidade de pagar novamente o mesmo valor se fosse comprar de novo um carro, pelo compromisso afetivo em ser cliente da concession?ria, pelo compromisso calculado quanto ?s perdas econ?micas de ter comprado este carro deste fabricante, em vez de outro similar de outro fabricante, e pelo tratamento das reclama??es

Submitted by Automa??o e Estat?stica (sst@bczm.ufrn.br) on 2017-07-04T14:52:40Z No. of bitstreams: 1 HelidaMariaBezerra_DISSERT.pdf: 9931928 bytes, checksum: 26e1ff7ff2ea69cdafbbf8e0db826dbe (MD5) / Approved for entry into archive by Arlan Eloi Leite Silva (eloihistoriador@yahoo.com.br) on 2017-07-14T14:22:38Z (GMT) No. of bitstreams: 1 HelidaMariaBezerra_DISSERT.pdf: 9931928 bytes, checksum: 26e1ff7ff2ea69cdafbbf8e0db826dbe (MD5) / Made available in DSpace on 2017-07-14T14:22:38Z (GMT). No. of bitstreams: 1 HelidaMariaBezerra_DISSERT.pdf: 9931928 bytes, checksum: 26e1ff7ff2ea69cdafbbf8e0db826dbe (MD5) Previous issue date: 2016-10-07 / Objetivo: Aumentar a ades?o ?s pr?ticas de preven??o de pneumonias associadas ? ventila??o mec?nica (PAV) em unidades de terapia intensiva (UTI), atrav?s de um ciclo de melhoria interno. Metodologia: Foi aplicado um ciclo de melhoria da qualidade com desenho quase experimental do tipo antes e depois em duas UTI de um hospital p?blico no nordeste do Brasil, com um total de 19 leitos. Foram estabelecidos nove crit?rios de qualidade relacionados com a preven??o de PAV e um indicador composto com os quais se avaliou o n?vel basal de qualidade e posteriormente mais duas avalia??es subsequentes com intervalos de quatro meses buscando medir as poss?veis melhorias adquiridas nesses per?odos. O Gr?fico de Pareto foi utilizado para representar a frequ?ncia de n?o cumprimentos de cada crit?rio avaliado nos tr?s momentos. Utilizou-se o c?lculo de estima??o pontual e o intervalo de confian?a (95%) para a medi??o dos crit?rios. Para quantificar a efetividade da interven??o, calculou-se a melhoria absoluta e relativa, al?m da significa??o estat?stica da melhoria absoluta mediante o teste unilateral do valor de z. Resultados: Ocorreu uma ligeira melhoria em todos os crit?rios quando comparadas as duas primeiras avalia??es, por?m nesta compara??o inicial, apenas um crit?rio apresentou melhora estatisticamente significante (p<0,05). Na segunda compara??o feita entre a 1? e 3? avalia??es, os resultados revelaram melhora no cumprimento de quase todos os crit?rios, onde dos nove crit?rios estabelecidos, oito obtiveram melhora, sendo seis com melhora estatisticamente significante (p < 0,05), dois com melhora sem signific?ncia e em um crit?rio houve piora no percentual de cumprimento. O indicador composto, que resume todos os crit?rios avaliados, obteve uma melhora significativa de quase 40% (p<0,05) na segunda compara??o. Conclus?es: A implanta??o de ciclos internos de melhoria constitui-se em importante ferramenta de gest?o da qualidade para incorpora??o de boas pr?ticas para preven??o de PAV em UTI. / Objective: To increase adherence of the preventive practices concerning Ventilator-associated pneumonia (VAP) in intensive care units (ICU), through an internal improvement cycle. Methodology: A quality improvement cycle, quasi-experimental design before-after was applied in two ICUs of a public hospital in the northeast of Brazil, with a total of 19 beds. Nine quality criteria related to the prevention of VAP were established and a composite indicator with which to assess the baseline level of quality and later two more subsequent assessments at intervals of four months seeking measure possible improvements acquired during these periods. The Pareto chart was used to represent the frequency of non-compliance of each criterion assessed in three moments. It was used the point estimation and the confidence interval (95%) for the measurement of the criteria. To quantify the effectiveness of the intervention, it calculated the absolute and relative improvement, beyond the statistical significance of the absolute improvement through unilateral test value of z. Results: There was a slight improvement on all criteria when comparing the first two assessments, but this initial comparison, only one criteria showed a statistically significant improvement (p <0.05). In the second comparison between the 1st and 3rd evaluations, the results showed improvement in the performance of all the criteria almost, where the nine established criteria, eight showed improvement, and six with a statistically significant improvement (p <0.05), with two improvement without significance and one criteria has worsened in the percentage of compliance. The composite indicator, which summarizes all the criteria evaluated, obtained a significant improvement of almost 40% (p<0.05) in the second comparison. Conclusions: The implementation of an internal improvement cycle was an important tool in quality management for the adoption of best practices for the prevention of VAP in ICU.