Appendix 1

Studies on perception of safety by residents of developed countries.

Source / Study object / Methodology / Sample size (Res-ponse rate) / Sample origin / Results / Conclusions / Recommendations
Experiences and perceptions / Daniel et al. 1999 / Experience of patients who filed a complaint / Participants were sent by mail a 32-item questionnaire which asked for a description of the incident leading to the complaint. / n = 500
(63%) / Patients with complaints managed by Health Care Complaints Commission (HCCC), in New South Wales (Australia) / A total of 64% of patients complained of the healthcare; 22% of how they were treated and 14% of lack of ethics. A total of 37% of the complaints were rejected, 2% had unknown results. Some 40% of patients thought the physician had been sanctioned. / Although patients believed they were doing the right thing by filing the complaint, most were not satisfied with the results and expected more convincing actions. Most wanted a punishment or a disciplinary action for the physician. Although complaints for clinical incidents are less common than those related to communication or personal ethics problems, the former are more likely to end in litigation. / Further knowledge of what people consider to be a lack of good medical practice is needed. Also, to understand patients’ expectations and to assess what they can legitimately accomplish is required.
Kaiser Family Foundation 2000 / AE experiences and opinions. / Telephone survey for patients and mail or online survey for physicians.
Firstly, respondents were asked about satisfaction with health care system, two mains problems of health care, and their knowledge about AE meaning. After, a definition of medical error was given, and they were asked closed-ended questions regarding preventable AE. These questions explored perceived frequency of AE, deaths due to them, causes of serious AE (either individual professionals or institutions) and a ranking of AE, effectiveness of possible solutions, confidentiality vs release of reports of AE, disclosure of AE, and safety depending on the hospital size. Also, two vignettes depicting a medication error with good and bad outcomes were presented. Then, they were asked about experiences of AE. / n= 2038
(831 physicians, and 1207 members of the public) / American public and physicians / 42% of the public had experienced an AE either in their own care or in a relative care. A total of 89% support that physicians should disclosure a serious AE. The public rated as the most effective solutions to prevent AE the following: spending more time with the physician (78%), requiring hospitals to develop systems to avoid AE (74%), improving training of health professionals (73%), and use only doctors specially trained in intensive care medicine or units (73%).Some / If information regarding quality levels of healthcare services or plans was available, patients would base their choice of medical insurance on that information. / The physician’s level of qualification is key to determining health care quality.
Adams and Boscarino 2004 / AE experienced in the household along their lives and in the last 5 years / Telephone survey
Research questions were: ‘How many persons reported that someone in their household experienced a medical error?’, ‘What were the key correlates of these errors?’, ‘Have you or anyone in your house hold ever suffered injury or harm that resulted from a medical error?’ Respondents were asked if the household AE occurred < 5 years ago, and about the type of error (medication, surgical, or diagnostic).
The authors focused particularly on whether being more informed about health care was related to a lower risk AE. / n= 1001
(53%) / Adult population living in the state of NY (US) / A total of 21% of the residents reported an AE had suffered by themselves or someone in their household (95%CI 18.3-24.0).
11.4% of the households had experienced an AE in the past 5 years (95%CI 9.2-13.6). / In 1/5 of New York homes people had suffered an AE, 1/10 in the last 5 years. The likelihood of informing of an AE would increase if people had greater knowledge of healthcare. / It is important to establish AE predictors in clinical histories or ones reported by patients in surveys, and also to analyze the specific features of the AE experienced by the people surveyed and their relatives.
Kuzel et al. 2004 / Main AE, types of AE and related harm in primary care / In-depth interviews.
The authors solicited narratives of preventable incidents in primary care that resulted in a perceived harm, based in a given broad definition or error. Respondents were asked to describe both the incident and the harm with as many additional stories as they could recall, helped by a cue card listing steps in primary care. / n=38 inter-
views / Residents in Virginia and Ohio
(US) / The 38 narratives related 221 incidents. 170 of them were linked to specific harms.
82.4% of the incidents involved breakdowns in the clinician-patient relationship, and 63.3% in access to physicians.
The majority of the referred harms were psychological (70%). / Patients’ accounts focus on psychological and emotional harm caused by medical errors, contrary to the current concern about pharmacological and surgical AE. Errors patients were related to access difficulties and deterioration of physician-patient relationship, rather than technical errors in diagnosis and treatment. / System flaws must be explored and fixed. It is also necessary to reform medical education, healthcare financing and also to change the way in which healthcare-related harm is dealt with, so as not to damage the physician-patient relationship.
Vander-heyden et al. 2005 / Perceptions regarding avoidable AE and personal and relatives’ experiences along their lives. / Phone survey, with content analysis for open questions.
Closed-ended questions elicited perceptions of preventable AE. Respondents were asked if they or a family member had ever experienced a preventable AE. Further closed-ended questions explored health consequences, and persons or institutions responsible of the error. Open-ended questions were: ‘What was the error?’, ‘What do you think caused the error?’, ‘How could the error have been prevented?’ / n= 1500
(55%) / Residents in Alberta (Canada) / A total of 37.3% of citiziens reported a preventable AE experienced in their own or family member’s care. / People blame physicians first and after the system. They are more concerned about the process in which the AE occurs than about the AE itself. They advocate a patient-centered health care system and think that coordination between physicians should be strengthened in order to improve health care quality and reduce the incidence of AE. / Diverse perspectives should be combined in order to understand avoidable AE and promote initiatives on patient safety. The patient’s perspective is to play a key role in research studies and proposals in order to gain his/ her trust and safety.
Agoritsas et al. 2005 / AE frequency reported by recently discharged patients and correlates of AE. / Mail survey
Two items were taken from the Picker patient opinion instrument, regarding if the patient had felt treated with respect and dignity during the hospital stay, and the evaluation about global rating of hospital care. The authors developed a list of 27 AE that are noticeable to patients and that occur commonly in a general hospital, including interpersonal problems, medical complications, and health care process problems. / n = 2156
(70%) / Patients discharged from a UniversityHospital (Switzerland) / Slightly over half of the respondents (50.6%) reported at least one AE.
There was an association of dissatisfaction with interpersonal problems (OR=1.6 CI 95% 1.3-1.8) and process-related problems (OR=1.5 CI 95% 1.3-1.9). / Patients can report AE occurring during the healthcare in the hospital. These AE occur in half the hospitalizations and produces a negative impact on satisfaction. / Patient AE notification can be a quantitative indicator of quality and safety.
Burroughs et al.2005 / Perception of safety when medical errors occur. / Phone interviews.
The main questions addressed were: ‘How safe do patients in Emergency Department (ED) feel from medical errors, and what types of errors are of greatest concern?’, ‘How are these concerns related to patient and hospital characteristics?, ‘What is the relationship between patient concern about medical errors and outcomes such as satisfaction and willingness to return for future care?’ Patients were asked about 23 items assessing satisfaction with an ED experience. Also, they were asked to rate the overall level of medical safety perceived during their ED stay on a five-point scale, and whether there was a specific time during their ED stay they were concerned about 8 possible AE. / n= 1262
(61%) / Patients from 12 Emergency Departments (US) / A total of 88% of the patients considered safety to be satisfactory when faced with medical errors. A total of The perception of safety was correlated with the wish to return to or recommend the ED (R2=0.57, p<0.0001). / Most people felt safe from errors. Concern is related to some kind of specific fear, not to errors in general. The type and number of concerns varied depending on patient and hospital characteristics and were closely related to satisfaction. The intention of returning to the ED or to recommending it to their relatives or friends was correlated with safety perception and concern. / Qualitative studies need to be carried out in order to analyze these patients’ fears and concerns. It is important to conduct research on the correlation between errors and patient perception.
Eurobarometrer 2006 / Perception of safety in European hospitals
Experiences of AE and practical implications / Home interviews
Respondents were asked closed-ended questions about the following: knowledge regarding AE and their importance, confidence in health professionals, experience of AE, worry about suffering a medical error, perception of safety in hospitals, likelihood of preventing AE, / 24,642 personal interviews / Population of the 25 countries belonging to the EU and of countries which are candidates for EU membership. / A total of 18% of those interviewed have been (either themselves or their relatives) the victims of a AE during their stay in a hospital. Latvia, Denmark and Poland lead the list of countries with greater hospital errors. / One out of 10 Europeans worries about being the victim of a medical error when he/ she goes to a hospital. / Clinical perceptions of safety vary considerably from one country to the other.
Evans et al. 2006 / Public opinion on the perception of safety in hospitals, and the rate and severity of AE occurring in these settings in the last 5 years, concluding relatives’ experiences of AE. / Telephone survey.
Respondents were asked, ‘In the last 5 years, how many times have you and members of your current household been admitted to hospital?’ If there had been a hospitalization, they were asked, ‘With regard to those hospitals, did anything ever go wrong that you might have been due to the way the health care was carried out?’ If yes, they were directed to rate the severity of the AE on a three-point Likert scale, and where they thought the AE prolonged the hospital stay. / n= 2884 homes, representing 8068 people
(78%) / Adult population living in the south of Australia / A total of 67% of respondents who were over 40 years old, had at least 1 person living in the same home hospitalized in the last 5 years, with an average of 2 hospitalizations. Some 7% of these were related to AE (CI 95% 6.2%-7.9%).
A total of 59.7% of respondents rated their AE as really serious (CI 95% 51.4%-67.5%), and 48.5% stated that the AE prolonged hospital stay (CI 95% 40.4%-56.8%). Very serious AE were the main cause of lack of perception of safety in patients who were 40 or older (RR= 2.38; (95% CI 1.85-3.07%).
51.6% of respondents (95% CI 49.8%-53.5%) would feel pretty safe. / Having suffered an AE had a negative impact on patients’ trust in the hospital; most of the interviewed people rated their AE as being very serious. The rate of AE occurring in hospitals estimated by those interviewed is similar to the rate identified in clinical histories. / Patients’ claims should be considered when developing appropriate treatment regimes. For patients who feel unsafe in hospitals, specific strategies could be designed, such as orientation prior to hospitalization and early discharge with home care.
Peters et al. 2006 / Influence of concern and perceived risk on preventive decision-making when faced with medical errors / Survey.
Participants were asked to estimate the number of deaths per year in the US from a variety of causes, including medical errors.They were asked how worried they were about each cause of death (from 0 to 6). After, participants rated 14 actions to prevent AE, and they were asked to evaluate 29 possible AE at hospital, in 6 scales of risk. Also, both half of the participants were asked to rate these AE on either 2 or 3 different scales, in order to evaluate Worry and Risk Likelihood. In addition, respondents who rated 3 behavioral intention items, were asked:‘How necessary do you believe it is to have new government regulations to reduce the occurrence of medical errors?’ Finally, they responded to a number of items assessing their reactivity to negative events, according to the Behavioral Inhibition System scale. / n=195
(100%) / College graduates of University of Oregon (US) / Average of concern = 3 (on a 0-6 scale). Average of deaths caused by errors estimated by patients= 44,022; median= 8025. The more concerned patients are, the more willing they are to take preventive actions before going to the hospital or while staying there, as well as to supporting a government regulation (r=0.41, 0.43 and 0.42, respectively, p<0.001) / The number of deaths caused by AE was underestimated. Patient-estimated deaths caused by errors were correlated with the errors they are most concerned about. Personal experience of AE was not related either to concern about errors or to risk perception. Concern led them to take preventive actions. / It would be more effective to focus error prevention models on factors based on feelings such as fear or negative reactivity rather than focusing them on purely cognitive factors. More research is needed in order to have patients as “surveillance partners” in the healthcare setting.
Disclosure of AE. Experiences and perceptions / Vincent et al. 1993 / Psychological impact of surgical accidents.
Adequacy of explanations given to the patients. / Mail survey.
The respondents were asked for a brief description of the incident, the effects the incident had on their live, explanations given before and afterwards the incident, whether patients had received an admission of responsibility or apology, whether they felt the incident was preventable and, if so, who they blame, and whether they had considered litigation and their reasons for proceeding or not.
General Health Questionnaire (GHQ-28), McGill pain questionnaire and the psychosocial adjustment to illness scale were adopted for this study. / n=154
(66%) / Patients who had contacted an organization which provides support for patients injured during their treatment
(AVMA).
Reino Unido / 66% of the patients rated the overall accident effect as severe or very severe.
Patients were dissatisfied with the explanations they were given. The poorer explanations were associated with a more painful memory after one year (R=0.32 CI 95% 0.16-0.49, p<0.01) and a worse social, workplace and disease adaptation (R=0.31, CI 95% 0.14-0.48, p<0.01, R=0.35, CI 95% 0.19-0.51, and R=0.36 CI 95% 0.20-0.52, p<0.001, respectively) A total of 50% showed less trust in physician competence and 45% did so in personal aspects. / Surgical accidents have a great psychological impact on patients; and this impact could be worse with subsequent problems in communication. / Physicians should inform patients of what happened in a simple and concise way, providing them with enough time to assimilate the information and to ask questions. It is convenient to provide some patients with psychological or psychiatric support, and also to advise and support the physicians involved in these cases.
Professionals require specific education to handle AE. Health authorities should establish a Code of Conduct to treat patients who have been harmed to help patients so that they do not need to take the matter to court.
Witman et al. 1996 / Attitudes when informing a patient of a medical error / Mail survey.
Participants were presented 3 scenarios involving AE of increasing severity. They were asked: ‘What type of response do the patients expect from the physician?’, ‘With whom would they like to further discuss the incident?’, ‘What would their response be?’, ‘How would that response been if the physician did not tell them of the mistake but they discovered it by some other means?’ / n=400
(37%) / Patients from a General Internal Medicine outpatient clinic in a university medical center (US) / A total of 98% of the people surveyed were in favor of the acceptance of even mild errors. Some 65% would see another physician after a serious AE. When faced with a moderate error, 20% of patients would file a complaint if his/her physician hadn't informed them of the error and if the patient had found it out by other means. On the other hand, if the physician had informed them of the error, only 12% of the patients would file a complaint (p<0.001) / Patients want physicians to admit even mild errors, which would reduce the risk of punitive actions. These findings support the idea that open communication between the physician and the patient is important. / Physicians should go on acting as patient advocates, but admitting honestly that they are not infallible. On the other hand, the profession needs to play a strong and creative role in order to reduce the frequency of errors. This strategy may reduce the risk of disciplinary actions against physicians.
Moore et al. 2000 / Causal effects of the physician-patient relationship and of the severity of the consequences on patient perception and intention of filing a complaint due to a medical error / Patients were sent by mail 1 of 4 scenarios depicting interactions between an obstetric patient and her physician throughout the patient’s pregnancy, labor and delivery. Participants were asked to complete the standardized questionnaire, ‘Patient-Doctor Interactions Scale’.
Using a 5-point scale they were asked to indicate their beliefs about the competency of the physician, the predictability of the complications, and the physician’s responsibility. They were also asked to state the percentage of responsibility (from 0%-100%) attributed to the physician, the nursing staff, the patient herself and chance. Finally, participants were asked to indicate, using a 5-point scale, the likelihood that they would file a malpractice claim against the physician and/or the hospital. / n=128
(81%) / Patients who had just given birth belonging to the University of California School of Obstetrics and Gynaecology, San Francisco (US) / Patients who communicated well with their physician were more satisfied (average=3.8 on a scale of 5) than those who didn’t have a good communication (average=1.8; F1,103=291.4;p<0.001), they perceived a better physician competence that the latter (average=3.1, 2.3, respectively; F1,1=8.13; p<0.01) and they had less intention of suing the physician for malpractice (F1,102=22, p<0.001) and of suing the hospital (F1,101=5.61; p<0.05). The intention to file a complaint against the hospital only increased when more serious AE occurred (F1,101=8.6; p<0.01). / A good physician-patient relationship improves the perception of physician competence, diminishes the attribution of the undesired outcome to the physician as well as the intention of filing a complaint against the physician or the hospital. Such intention only increases when serious errors occur. / Future studies should examine the effects of communication-specific behavior of physicians and patients. This information could reduce stressing both groups, potential legal actions and costs related.