A survey on patient safety culture in primary healthcare services in Turkey

Transcription

1 International Journal for Quality in Health Care 2009; Volume 21, Number 5: pp Advance Access Publication: 22 August 2009 A survey on patient safety culture in primary healthcare services in Turkey SAID BODUR 1 AND EMEL FILIZ /intqhc/mzp035 1 Department of Public Health, Meram Faculty of Medicine, Selcuk University, Konya, Turkey, and 2 Public Health Nurse, College of Health, Selcuk University, Konya, Turkey Abstract Objective. To evaluate the patient safety culture in primary healthcare units. Design. A cross-sectional study, utilizing the Turkish version of the Hospital Survey on Patient Safety Culture developed by the Agency for Healthcare Research and Quality and a demographic questionnaire. Setting. Twelve primary healthcare centers in the center of the city of Konya, Turkey. Participants. One hundred and eighty healthcare staff, including general practitioners (GPs), nurses, midwives and health officers. Intervention. None. Main Outcome Measure(s). The patient safety culture score including subscores on 12 dimensions and 42 items; patient safety grade and number of events reported. Results. Fifty-four (30%) of the participants were GPs, 48 (27%) were nurses, 51 (28%) were midwives and 27 (15%) were health officers. The mean overall score for positive perception of patient safety culture in primary healthcare units was (43 49 CI). No differences were found by staff members profession. Among the dimensions of patient safety, those with the highest percentage of positive ratings were teamwork within units (76%) and overall perceptions of safety (59%), whereas those with the lowest percentage of positive ratings were the frequency of event reporting (12%) and non-punitive response to error (18%). Reporting of errors was infrequent with 87% of GPs, 92% of nurses and 91% of other health staff indicating that they did not report or provide feedback about errors. Conclusions. Improving patient safety culture should be a priority among health center administrators. Healthcare staff should be encouraged to report errors without fear of punitive action. Keywords: patient safety, patient safety culture, primary healthcare unit, general practitioner, nurse Introduction Patient safety is defined as the prevention of harm caused by errors of commission and omission [1]. One-fifth of the people in the community are exposed to medical mistakes [2], and this rate may be as high as 35 42% [3]. As a result, millions of people may die or suffer injuries due to preventable medical errors. The widespread nature and heavy consequences of medical mistakes require more studies focusing on patient safety [4]. These types of studies generally concentrate on hospital environments [5]. The area of primary health care concerns everyone in the community because it provides the first contact for the patient. However, since severe and complicated cases requiring special treatment are handled in hospitals, both providers and the community frequently underestimate the importance of primary healthcare services. This underestimation leads to a primary care environment susceptible to errors in fields such as organization, physician notification, communication and staffing [6]. Thus, some studies [7, 8] have found that errors in primary care can result in serious consequences. Primary healthcare units need an organizational safety culture similar to that established in hospitals [9, 10]. An institutional culture involves the procedural flow in a given institution; a safety culture is one in which safety is everyone s concern [10]. In this regard, patient safety culture can be defined as acceptance and practice of patient safety as the Address reprint requests to: Said Bodur, Meram Tip Fakültesi, Halk Sagligi Anabilim Dalı, Selcuk Üniversitesi, Konya, Türkiye. Tel: þ ; Fax: þ ; saidbodur@gmail.com International Journal for Quality in Health Care vol. 21 no. 5 # The Author Published by Oxford University Press in association with the International Society for Quality in Health Care; all rights reserved 348

2 Patient safety culture in Turkey first priority and the common value in the institution. In other words, patient safety culture may be described as the common values, beliefs, behaviors, perceptions and attitudes of the staff in a healthcare center [11]. The generation of a safety culture starts with an evaluation of the present safety level in an institution because safety precautions implemented without a proper assessment may elevate costs and also causing unpredicted new risks [12]. Many tools have been developed for evaluation of the patient safety culture [13 15]. Nearly all these tools cover five common dimensions of patient safety climate: leadership, policies and procedures, staffing, communication and reporting [16]. Patient safety culture is a relatively new area [12], and most of the studies published in this field are based on studies of hospitals [17]. There are few studies of patient safety culture in primary healthcare services and none conducted previously in Turkey. Also, the number of valid and reliable surveys related to patient safety culture for primary healthcare services worldwide is limited. For this reason, a modification of the Hospital Survey on Patient Safety Culture (HSOPSC) has been used in non-hospital settings such as nursing homes. In this research, we analyze the responses of primary healthcare service staff to a modified version of the HSOPSC. The purpose of this study was to determine safety culture scores for primary healthcare services and compare these results with existing data from benchmark scores using HSOPSC. The differences between different types of professional staff (e.g. physicians, nurses and others) were also assessed. Methods Setting Twelve primary healthcare services were randomly selected from 37 primary healthcare services in the metropolitan city center of Konya, Turkey. Primary health care, including outpatient care, maternal care, family planning, immunization and preventive health services are provided through a network of primary healthcare services that were established throughout the country on the basis of a 1963 law regarding socialization of health services. Primary healthcare services were configured for rural and urban areas. Each center of rural primary healthcare services was expected to serve 5000 people. In urban areas, as in the present study, each center is expected to serve people. Primary healthcare services are to be staffed by a team of four to six general practitioners (GPs), a nurse or midwife for each 3000 people, several health officers and other ancillary staff [18]. Participants The data were obtained from a self-administered questionnaire survey completed in The sample size was determined according to a formula based on a comparison of the predicted mean score [19] of the positive perception for the patient safety culture dimensions, setting the sample size so that differences between groups would be detected with a power of 80% at a 95% confidence level within a relative deviation of 10%. Thus, a sample size of at least 44 people for each of the four groups of healthcare professionals was planned. Questionnaires and informed consent forms were hand-distributed to 212 staff. Of this group, 185 persons who gave consent completed the survey (response rate 85%). The numbers of the four types of healthcare staffs included in the study were: 54 GPs, 48 nurses, 51 midwives and 27 health officers. GPs are responsible for preventive medicine (immunization, monitoring of child and mother, health education etc.) and outpatient care. Nurses have the responsibility for immunization, family planning, patient care, injections and dressings. Midwives follow pregnant women and infants, as well as post-partum women, and aid in immunization and family planning. Health officers are involved with bureaucratic management. Measures The survey included a section that asked questions from the HSOPSC. It is difficult to find a universal culture instrument in the literature on current assessments of patient safety culture [16]. This is especially so in the case of outpatient clinics such as primary healthcare services. Therefore, we used the HSOPSC developed by the Agency for Healthcare Research and Quality (AHRQ) [20]. This instrument contains 12 subscales and 42 items that consider many attributes known to be associated with a culture of patient safety, identified above [21]. Specifically, the subscales of the instrument include: (i) manager expectations and actions promoting safety; (ii) organizational learning; (iii) teamwork within units; (iv) communication openness; (v) feedback and communication about errors; (vi) non-punitive response to errors; (viii) staffing; (viii) management support for patient safety; (ix) teamwork across units and (x) handoffs and transitions. The HSOPSC also includes two subscales that are presented as outcome dimensions: (i) overall perceptions of safety and (ii) frequency of event reporting. The HSOPSC is a valid and reliable instrument developed from previous literature, cognitive tests and factor analyses to assess the patient safety culture in hospitals. Extensive details of this instrument can be found in a web-based technical report [21]. The final instrument was pilot-tested in 21 hospitals with 1437 employee responses. Using Cronbach s a, all subscales had acceptable levels of reliability, which varied from 0.84 for frequency of event reporting to 0.63 for staffing. The construct validities of each safety culture dimension were shown in composite scores as being moderately related to one another, as indicated by correlations between 0.20 and 0.40 [21]. Preparing a Turkish form of the AHRQ survey Initial translation of the survey into Turkish was performed by the investigators. This translation was reviewed by an English-language expert whose native language is Turkish. It was then translated back into English by an independent translator who had not seen the original questionnaire [22]. 349

3 Bodur and Filiz This form was examined by one of the authors of the original HSOPSC to determine whether different wordings in the new survey text in English and the original one posed any problems and whether it could be used in its new form. The authors made revisions after the consultant suggested that several items in the survey could lead to different interpretations. Those items were adjusted according to the comments of consultant. As we used this instrument in primary healthcare services, a modification was made to the wording of items. Where it made sense, the word hospital was replaced by primary healthcare services. No other changes were made to the instrument. The format, response options and question order remained the same as in the original instrument. Data analysis Descriptive statistics for the facilities in the sample and descriptive statistics for each item on the HSOPSC were calculated. For each positively worded item, the percentage of positive responses was calculated, i.e. the percentage of respondents answering the question as strongly agree and agree or always or most of the time. Similarly, for each negatively worded item, the percentage of negative responses was calculated. In addition, the mean for each subscale used (listed above) in the HSOPSC was calculated. Subscale scores were calculated by taking the average score of the subscale items. In all cases, the possible range of scores is from 0 to 100%, with higher scores indicating a more positive response. We used t-tests to compare the primary healthcare services item scores and subscale scores with the matching scores of 58 US hospitals with 6 24 bed-size (benchmark score) [17]. This questionnaire has not been used previously in primary healthcare services, so we also calculated Cronbach s a for each subscale. The HSOPSC instrument also includes questions asking respondents to give an overall grade to their safety environment and how many incident reports have been generated in the past 12 months. The responses to these questions are summarized below. Composite scores obtained from the study on each of the 12 safety culture dimensions and 42 items were compared with the benchmark scores using a t-test. The relationship between demography factors and patient safety culture score was examined using a linear multivariate analysis with mean percentage reporting a positive score as the dependent variable. A chi-squared test was used to compare health professions on each of the 42 safety culture items, on the two single-item outcome measures (number of events reported and patient safety grade), and on composite scores of the safety culture scales. Results In total, 180 healthcare staff members provided survey feedback (response rate 85%). Fifty-four (30%) of the participants were GPs, 48 (27%) were nurses, 51 (28%) were midwives and 27 (15%) were health officers. Sixty percent (60%) of the nurses and 56% of other staff had pre-bachelor s degree. The mean age of the participants was , and 89% of them were married. The majority of the healthcare staff had a seniority level above 15 years (Table 1). The percentage of positive responses was highest for teamwork within units (76%), overall perceptions of safety (59%) and teamwork across hospital units (56%) and lowest for frequency of event reporting (12%) and non-punitive response to error (18%). The overall mean score for positive perception of patient safety culture was (95% CI: 43 49), whereas it was for GPs, for nurses, for midwives and for health officers (P ¼ 0.054). Comparison of scores for the dimensions of safety culture revealed lower results for nurses compared with those of GPs and other healthcare staff in terms of communication openness (P, 0.01) and feedback and communication about error (P, 0.01) (Table 2). In the multivariate analysis, staff who had been working more than 10 years in their present unit displayed a significantly lower patient safety culture score (P, 0.05). Similarly, a weak but significant negative and linear correlation was Table 1 Sociodemographic and professional characteristics of health staffs a Characteristics GPs (n ¼ 54) Nurses (n ¼ 48) Midwives (n ¼ 51) Health officers (n ¼ 27)... Gender Male 24 (44) 0 (0) 0 (0) 27 (100) Female 30 (56) 48 (100) 51 (100) 0 (0) Age (years) 24 2 (4) 1 (2) 1 (2) 0 (0) (30) 27 (56) 31 (61) 14 (52) (53) 19 (40) 16 (31) 9 (33) 45 7 (13) 1 (2) 3 (6) 4 (15) Marital status Single 5 (9) 5 (10) 5 (10) 3 (11) Married 49 (91) b 43 (90) 46 (90) 24 (89) Work experience (years),5 3 (6) 2 (4) 3 (6) 2 (7) (28) 7 (15) 7 (14) 8 (30) (24) 19 (39) 21 (41) 8 (30) (42) 20 (42) 20 (39) 9 (33) Years in the facility,1 5 (9) 3 (6) 9 (18) 2 (7) (37) 25 (52) 19 (37) 15 (55) (30) 12 (25) 11 (22) 5 (19) (24) 8 (17) 12 (23) 5 (19) Hours of work per week (63) 36 (75) 40 (78) 22 (81) (37) 12 (25) 11 (22) 5 (19) a Parenthesis refers percentage. b Two of widowed. 350

4 Patient safety culture in Turkey Table 2 Mean patient safety culture composite scores across professions Patient safety culture dimension GPs (n ¼ 54) Nurses (n ¼ 48) Midwives (n ¼ 51) Health officers (n ¼ 27) P-value Overall Perceptions of Patient Safety NS Frequency of Events Reported NS Mgr Expectations & Actions Promoting Patient Safety NS Org Learning Continuous Improvement NS Teamwork Within Units NS Communication Openness 56 32* 45 51,0.01 Feedback & Communication About Error 65 35* 53 46,0.01 Non-punitive Response to Error NS Staffing NS Management Support for Patient Safety NS Teamwork Across Units NS Handoffs & Transitions NS Overall NS NS, not significant. *Lower than GPs (using Tukey s HSD P, 0.05). Figure 1 Comparison of mean patient safety culture composite scores of Turkish primary healthcare services and benchmark scores [17]. (Obtained from 58 US hospitals having a bed size between 6 and 24.) 351

5 Bodur and Filiz Table 3 Descriptive statistics of survey on patient safety culture and benchmark scores Subscales and survey items Average % positive response a (SD) Benchmark score a Overall perceptions of safety (Cronbach s a ¼ 0.43) 59 (8) 69* Patient safety is never sacrificed to get more work done 60 (10) 66 Our procedures and systems are good at preventing errors from happening 67 (11) 72 It is just by chance that more serious mistakes do not happen around here 52 (8) 69* We have patient safety problems in this facility (r) 44 (13) 70* Frequency of events reported (Cronbach s a ¼ 0.93) 12 (8) 64* When a mistake is made, but is caught and corrected before affecting the patient, how 10 (7) 54* often is this reported? When a mistake is made, but has no potential to harm the patient, how often is this 12 (9) 60* reported? When a mistake is made that could harm the patient, but does not, how often is this 15 (12) 77* reported? Manager expectations and actions promoting patient safety (Cronbach s a ¼ 0.67) 58 (8) 76* Manager says a good word when he/she sees a job done according to established 58 (13) 72* patient safety procedures Manager seriously considers staff suggestions for improving patient safety 58 (11) 77* Whenever pressure builds up, my manager wants us to work faster, even if it means 38 (10) 79* taking shortcuts (R) My manager overlooks patient safety problems that happen over and over (R) 69 (8) 77* Organizational learning continuous improvement (Cronbach s a ¼ 0.56) 47 (11) 71* We are actively doing things to improve patient safety 50 (13) 83* Mistakes have led to positive changes here 44 (15) 65* After we make changes to improve patient safety, we evaluate their effectiveness 43 (15) 67* Teamwork within units (Cronbach s a ¼ 0.84) 76 (8) 83* People support one another in this facility 71 (15) 87* When a lot of work needs to be done quickly, we work together as a team to get the 76 (9) 90* work done In facility, people treat each other with respect 80 (10) 80 When one area in this unit gets really busy, others help out 70 (11) 73 Communication openness (Cronbach s a ¼ 0.67) 46 (7) 66* Staff will freely speak up if they see something that may negatively affect patient care 51 (11) 79* Staff feel free to question the decisions or actions of those with more authority 28 (10) 51* Staff are afraid to ask questions when something does not seem right (R) 52 (10) 66* Feedback and communication about error (Cronbach s a ¼ 0.82) 50 (11) 65* We are given feedback about changes put into place based on event reports 46 (16) 52 We are informed about errors that happen in the facility 57 (15) 68* In this facility, we discuss ways to prevent errors from happening again 47 (12) 74* Non-punitive response to error (Cronbach s a ¼ 0.13) 18 (6) 48* Staff feel like their mistakes are held against them (R) 9 (7) 55* When an event is reported, it feels like the person is being written up, not the problem 22 (12) 48* (R) Staff worry that mistakes they make are kept in their personnel file (R) 17 (9) 41* Staffing (Cronbach s a ¼ 0.02) 49 (5) 63* We have enough staff to handle the workload 57 (14) 64 Staff in this facility work longer hours than is best for patient care 35 (9) 58* We use more agency/temporary staff than is best for patient care 76 (11) 69 We work in crisis mode trying to do too much, too quickly (R) 16 (10) 59* Management support for patient safety (Cronbach s a ¼ 0.59) 42 (9) 75* Management provides a work climate that promotes patient safety 48 (15) 84* The actions of management show that patient safety is a top priority 33 (10) 75* Management seems interested in patient safety only after an adverse event happens 42 (14) 64* (continued ) 352

6 Patient safety culture in Turkey Table 3 Continued Subscales and survey items Average % positive response a (SD) Teamwork across units (Cronbach s a ¼ 0.73) 58 (6) 66* Units do not coordinate well with each other (R) 57 (8) 65* There is good cooperation among units that need to work together 65 (9) 55 It is often unpleasant to work with staff from other units (R) 49 (10) 77* Units work well together to provide the best care for patients 54 (10) 69* Handoffs and transitions (Cronbach s a ¼ 0.70) 44 (10) 56* Things fall between the cracks when transferring patients from one 43 (14) 55* unit to another (R) Important patient care information is often lost during shift changes (R) 40 (9) 57* Problems often occur in the exchange of information across units (R) 43 (13) 54* Shift changes are problematic for patients in this facility (R) 42 (12) 57* Overall 46 67* Benchmark score a a For each positively worded item, the percentage of positive responses was calculated i.e. the percentage of respondents answering the question as strongly agree and agree, or always or most of the time. R: item was reverse coded, so for all questions higher scores are more favorable. *Significantly different t-test at P, 0.05, using same SD for both primary data and benchmark scores. Nurses had lower mean scores than the others P, Table 4 Percentage of respondents giving their work area/unit a patient safety grade by professions Work area/unit patient safety grade* Physicians Nurses Midwives Health officers Overall** Benchmark Excellent Very good Acceptable Poor Failing *P ¼ **P, found between patient safety culture score and work years in the unit (r ¼ 20.21, P ¼ 0.011). No relationship was found between gender, occupation, marital status or weekly work hours and total patient safety culture score. Patient safety culture scores obtained from primary healthcare services were lower than the benchmark scores obtained from 58 US hospitals having a bed size between 6 and 24 (P, 0.001). The positive response scores of primary healthcare units for all 12 dimensions were lower than the benchmark scores (Fig. 1) [17]. The dimensions exhibiting the largest difference were as follows: the score for frequency of event reporting was 12% for primary healthcare centers, whereas it was 64% for hospitals, the score for non-punitive response to error was 18% for primary healthcare centers and 48% for hospitals, and the score for management support for patient safety was 43% for primary healthcare centers and 75% for hospitals (Graph 1). The assessment based on specific survey items showed that primary healthcare units scored lower than hospitals in 34 items out of 42. Cronbach s a values of the dimensions varied between 0.43 and 0.93 in 10 of 12 dimensions. Cronbach s a value was found to be below 0.40 in two dimensions ( non-punitive response to error and staffing ) (Table 3). The percentage of staff who rated the level of patient safety in primary healthcare units as good or perfect was 42%, which was lower than the benchmark score of 76% (P, 0.001) (Table 4). In addition, 87% of GPs, 92% of nurses and 91% of other healthcare staff were shown as never reporting the errors. So, the frequency of event reporting (10%) was very much lower than the benchmark score of 50% (P, 0.001) (Table 5). Discussion As far as we know, this study is the first one that investigated patient safety culture in primary healthcare services by healthcare providers. The instrument used to measure the patient safety culture was developed by the AHRQ for hospitals [21]. Use of this instrument in primary healthcare services is a limitation. Development of a new instrument aimed specifically at the primary healthcare services provided 353

7 Bodur and Filiz Table 5 Percent of respondents reporting events in the past 12 months by professions Number of events reported by respondents* Physicians Nurses Midwives Health officers Overall** Benchmark No events One to two events Three to five events Six events reports or more *P ¼ **P, by outpatient clinics and preventive medicine would be a better solution. In the present study, the response rate of the survey was comparable to that of previous studies. However, the study has some limitations. It was carried out only in the urban areas, excluding the rural primary healthcare institutions. Its scope should be further extended. A patient safety culture score is only an indirect indicator of patient safety. Practical efficacy may be obtained from the presented scores. However, greater number of sample populations should be used to obtain results applicable to more groups. For example, the opinions of the medical administrators and patients can be included in the study because this subject should have input from patients as well [23]. Physicians should be willing to assume the leadership required for the development of an institutional safety culture in healthcare services [24]. Moreover, the impact of patient safety culture interventions on personnel and patient outcomes may be evaluated by longitudinal studies. The current study is comprised 54 GPs and 126 healthcare staff who work in urban primary healthcare units delivering outpatient services. In two aspects ( communication openness and feedback and communication about error ), the scores of the nurses were lower than those of GPs, which was a remarkable finding. Nurses spend more time with the patients and can observe the relationship between the GPs and the patients. Therefore, lower scores for nurses in some aspects of safety culture may be regarded as especially salient. One of the demographic aspects, work years in a unit, stands out as a factor having an impact on patient safety culture score. Because patient safety culture scores decreased as seniority increased, one might wonder whether there has been an increase in medical errors over time, or whether the relatively lower score is related to some other mechanisms such as an increasing staff awareness of safety problems or greater willingness to admit to safety problems on a survey as staff gain more experience on a unit. Since there was no relationship between the patient safety culture score and other demographic dimensions, safety environment appears to be independent of other demographic attributes. The mean overall score of patient safety culture in primary healthcare units is lower than that of hospitals [17]. However, perception of patient safety culture by health staff members resulted in similar scores. Because primary healthcare institutions have a lower potential for life-threatening medical errors and procedures (because they do not manage severe clinical cases that can be treated only by intensive care, emergency medicine, major surgical interventions or inpatient services, we regard determination of the perception of patient safety culture by the healthcare staff as an important issue in terms of public health. As most risky medical interventions take place in hospitals, hospital staff may have better training and specialization in safety-related issues. This may account for the relatively lower reporting of problems regarding patient safety in hospitals compared with primary care. On the other hand, because the medical risk is estimated to be lower in primary healthcare units, patient safety precautions might be neglected or disregarded in those institutions due to the low risk potential, which may lead to the development of unexpected threats. Development of a positive safety culture in primary healthcare services requires further study based on results obtained from the current patient safety culture survey showing lower scores for primary healthcare units in all dimensions compared with the results for US hospitals [17]. Most of the mean values of primary healthcare services for patient safety culture were below 50%. However, positive perception was above 50% in the following dimensions: teamwork within units, supervisor/manager expectations and actions promoting safety, overall perceptions of safety and teamwork across units. Although the high values of primary healthcare services for overall perception of teamwork, cooperation and safety culture were encouraging, the lower frequency of positive responses in many fundamental aspects of safety culture was disappointing. Although the patient safety culture scores for primary healthcare services were lower than those for US hospitals [17], scores on some dimensions (overall perceptions of safety, teamwork within units, teamwork across units etc.) were higher than some hospitals in Belgium [25]. This difference may stem from inherent cultural differences or differences obtained from primary healthcare services or hospitals. Because efforts to identify mistakes may be undervalued in Turkish culture, reporting the errors seems to be a process that is often avoided. Turkey may need to improve the patient safety culture in primary healthcare services to reach the benchmark values for safety culture level and event reporting frequency. In the current study, the frequency of event reporting was much 354

8 Patient safety culture in Turkey lower than that in US hospitals [21]. Non-punitive responses to error and error-reporting should be improved. For instance, development of error reporting based on voluntary and consistent event reports [26] is recommended to improve patient safety in primary healthcare services. Our study suggests that the patient safety survey used in hospitals can be modified and employed in outpatient clinics. Some items could be revised according to primary care services for which inter-item reliability is low. In conclusion, the development of a patient safety culture should be a priority of administrators in primary healthcare units, as it is in hospital settings. An environment in which healthcare staff can report present or possible errors without fear of punishment should be established. References 1. Institute of Medicine (IOM). Patient Safety: Achieving a New Standard for Care. Washington, DC: National Academy Press, Adams RE, Boscarino JA. A community survey of medical errors in New York. Int J Qual Health Care 2004;16: Blendon RJ, DesRoches CM, Brodie M et al. Views of practicing physicians and the public on medical errors. N Engl J Med 2002;347: WHO. Call for More Research on Patient Safety, who.int/mediacentre/news/releases/2007/pr52/en/index.html. 5. Castle NG. Nurse Aides ratings of the resident safety culture in nursing homes. Int J Qual Health Care 2006;18: Kuzel AJ, Woolf SH, Gilchrist VJ et al. Patient reports of preventable problems and harms in primary health care. Ann Fam Med 2004;2: Dovey SM, Meyers DS, Phillips RL Jr et al. A preliminary taxonomy of medical errors in family practice. Qual Saf Health Care 2002;11: Rosser W, Dovey S, Bordman R et al. Medical errors in primary care: results of an international study of family practice. Can Fam Physician 2005;51: Institute of Medicine (IOM). Priority Areas for National Action: Transforming Health Care Quality., Washington, DC: National Academies Press, Milligan F, Dennis S. Building a safety culture. Nurs Stand 2005;20: Pizzi LT, Goldfarb NI, Nash DB. Promoting a culture of safety. Making Health Care Safer. AHRQ Publication No. O1-E058, 2001, Warburton RN. Patient safety how much is enough? Health Policy 2005;71: Schutz AL, Counte MA, Meurer S. Development of a patient safety culture measurement tool for ambulatory health care settings: analysis of content validity. Health Care Manag Sci 2007;10: Kirk S, Parker D, Claridge T et al. Patient safety culture in primary care: developing a theoretical framework for practical use. Qual Saf Health Care 2007;16: Emergency Care Research Institute (ECRI). Healthcare risk control: culture of safety. Risk and quality management strategies 21, Colla JB, Bracken AC, Kinney LM et al. Measuring patient safety climate: a review of surveys. Qual Saf Health Care 2005;14: Sorra J, Famolaro D, Dyer N et al. Hospital Survey on Patient Safety Culture: 2008 Comparative Database Report, Parts II and III. Rockville, MD: AHRQ Publication No , http :// 18. The Ministry of Health of Turkey. Turkey Health Report, Ankara, Sorra J, Nieva V, Famolaro T et al. Hospital Survey on Patient Safety Culture: 2007 Comparative Database Report. Rockville, MD: AHRQ Publication No , Nieva VF, Sorra J. Safety culture assessment: a tool for improving patient safety in health care organizations. Qual Saf Health Care 2003;12: Sorra JS, Nieva VF. Hospital Survey on Patient Safety Culture. Rockville, MD: AHRQ Publication No , http :// 22. Sperber AD. Translation and validation of study instruments for cross-cultural research. Gastroenterology 2004;126: Vincent CA, Coulter A. Patient safety: what about the patient? Qual Saf Health Care 2002;11: Classen DC, Kilbridge PM. The roles and responsibility of physicians to improve patient safety within health care delivery systems. Acad Med 2002;77: Hellings J, Schrooten W, Klazinga N et al. Challenging patient safety culture: survey results. Int J Health Care Qual Assur 2007;20: Fernald DH, Pace WD, Harris DM et al. Event reporting to a primary care patient safety reporting system: a report from the ASIPS Collaborative. Ann Fam Med 2004;2: Accepted for publication 20 July