OUP user menu

Culture, language, and patient safety: making the link

Megan-Jane Johnstone, Olga Kanitsaki
DOI: http://dx.doi.org/10.1093/intqhc/mzl039 383-388 First published online: 6 September 2006

Abstract

It has been well recognized internationally that hospitals are not as safe as they should be. In order to redress this situation, health care services around the world have turned their attention to strategically implementing robust patient safety and quality care programmes to identify circumstances that put patients at risk of harm and then acting to prevent or control those risks. Despite the progress that has been made in improving hospital safety in recent years, there is emerging evidence that patients of minority cultural and language backgrounds are disproportionately at risk of experiencing preventable adverse events while in hospital compared with mainstream patient groups. One reason for this is that patient safety programmes have tended to underestimate and understate the critical relationship that exists between culture, language, and the safety and quality of care of patients from minority racial, ethno-cultural, and language backgrounds. This article suggests that the failure to recognize the critical link between culture and language (of both the providers and recipients of health care) and patient safety stands as a ‘resident pathogen’ within the health care system that, if not addressed, unacceptably exposes patients from minority ethno-cultural and language backgrounds to preventable adverse events in hospital contexts. It is further suggested that in order to ensure that minority as well as majority patient interests in receiving safe and quality care are properly protected, the culture–language–patient-safety link needs to be formally recognized and the vulnerabilities of patients from minority cultural and language backgrounds explicitly identified and actively addressed in patient safety systems and processes.

  • culture
  • ethnic minorities
  • health care
  • language
  • patient safety
  • quality care

It is well recognized internationally that hospitals are not as safe as they should be. Over the past decade, studies conducted respectively in the United States of America, Australia, the United Kingdom, Europe, and Canada have found that between 4 and 16.6% of patients suffered some kind of harm (including permanent disability or death) while in hospital [1–5]. These studies further found that a large percentage (∼50%) of the harmful outcomes recorded could have been prevented, that is, were preventable adverse events. Concerned by these and other reports, governments, health care services, and health professional groups around the world turned their attention to strategically implementing robust patient safety and quality care programmes that have as their specific focus and emphasis identifying circumstances that put patients at risk of harm, and then acting to prevent or control those risks [6–9].

In the years since these reports were first made public, significant progress has been made in motivating understanding of the multifaceted processes contributing to preventable adverse events in health care domains and achieving the changes needed in organizational culture, systems, training, and technology to improve patient safety and related outcomes. Despite the progress that has been made in recent years, there remains a significant patient safety issue that has yet to be formally recognized and systematically addressed, namely, the issue of culture and language and its possible link to patient safety outcomes in patient groups of diverse minority cultural and language backgrounds. It is a key aim of this article to contribute to the positive project of redressing this ‘gap’ in contemporary patient safety discourse.

Influences of culture and language in health care

It is widely recognized in the international literature that culture has a significant influence on shaping people’s perceptions and experiences of health and health care and on how—whether as the providers or the recipients of care—they make sense of, give meaning to, and respond to their experiences [10–14]. It is also being increasingly recognized that a patient’s culture and language (including the way patients speak a mainstream language, e.g. English) are significant predictors of the quality of care delivered, with research strongly suggesting that patients of minority cultural and language backgrounds tend to receive poorer quality of care than do majority patients [10,14,15]. Less well recognized, however, is the influence that cultural and language variables (especially patient-provider cultural and language incongruence) may have on the incidence and impact of preventable adverse events in patients from minority racial, ethno-cultural, and language backgrounds [16].

Over the past decade, and in keeping with what has been described as a new ‘global agenda’ for patient safety in health care [17], there has been a plethora of literature published on the topic of patient safety. There is, however, a critical and unrecognized ‘gap’ in this literature concerning patient safety processes pertinent to reducing the vulnerabilities of patients from minority cultural and language backgrounds. As revealed by a recent library search of multiple electronic databases (for all years) using the key words ‘patient safety’, ‘clinical risk management’, ‘ethnic minorities’, ‘culture’, ‘language’, ‘health care’ (and variations thereof), there is a paucity of literature specifically addressing the critical relationship that exists between culture, language, and patient safety, and the particular risks that patients from minority racial, ethno-cultural, and language backgrounds face when being cared for by health care professionals who do not know about, share, or understand either their culture or language.

Although numerous articles (too numerous to cite here) have been published in which the term ‘culture’ is used, the term has been used primarily to refer to organizational culture [notably, as an ‘independent variable that can be manipulated through management interventions in order to achieve organizational goals’ ([18], p.345)], not the ‘human culture’ of patients and health professionals. The notion of ‘organizational culture’ and the discourses surrounding it are, however, limited. In the context of debates about culturally responsive health care, the notion of ‘organizational culture’ is unable to provide a substantive account of the complexities and dynamics of culture in its anthropological sense [i.e. of it being a ‘shared set of values, ideas, concepts, and rules of behaviour that allow a social group to function and perpetuate itself’ ([18], p.345)], and its significant influence on the clinical encounter (between the providers and recipients of health care) in hospital contexts. The notion of organizational culture is also insufficient to motivate either understanding or practice (whether in managerial, clinical, educational, or research domains) that is reflective of the critical link between culturally and linguistically responsive clinical encounters (culture care) and patient safety outcomes for patients of diverse cultural and language backgrounds.

Establishing the link between culture, language, and patient safety

There is an acknowledged lack of comparable data and ‘gold standard’ research demonstrating a conclusive link between cultural and language considerations in clinical contexts and patient safety outcomes. Researchers are, however, slowly turning their attention to demonstrating this link. For example, in what is believed to be the first systematic review of the international literature examining ‘the impact of medical interpreter services on the quality of health care’, Flores [19] found substantial evidence to support the conclusion that when patients of limited English proficiency (LEP) need but do not get the services of qualified health interpreters, their quality of care and related health outcomes are seriously compromised and, in some areas, demonstrably inferior compared with that experienced by other patients who have English language proficiency.

Other studies have had similar findings. For example, in a study investigating what, if any, link exists between ‘language barriers’ and ‘serious medical errors’ during paediatric hospitalizations, Cohen et al. [20] found that a subgroup of Spanish-speaking patients whose families did not speak English had a significantly increased risk of serious medical events compared with patients whose families did not have a language barrier. The researchers go on to point out that the risk of events was probably even higher than calculated on account of the fact that their study had not captured other less serious events and ‘near misses’ during the period studied and thus were underestimated ([20], p. 578).

Newly mandated patient-safety-reporting systems are also capturing data implicating cultural and language considerations (particularly the failure to use accredited health interpreter services) in preventable adverse events. In the Australian State of Victoria, for example, following the creation in 2003 of a new category to capture ‘ethnic data’ in mandated sentinel event reporting to the Victorian Government Department of Human Services, a number of incidents involving non-English-speaking patients who underwent wrong site procedures were reported [21,22]. In each of these cases, the patients concerned were not provided with access to interpreter services, which, ‘combined with insufficient site checking procedures, resulted in procedures being undertaken on incorrect body parts’ ([22], p. 24).

Legal cases and the culture, language, patient safety link

The issue of culture and language and the risks associated with ‘cultural misunderstandings’ and failures to use professional health interpreters in the clinical encounter have long been legally implicated in preventable adverse events in patients of minority cultural and language backgrounds.

In a 1984 US case, for example, the misinterpretation of one word during a medical consultation in a Florida Emergency Department lead to the misdiagnosis of a cerebral haemorrhage in a patient that resulted in a $71 million award for damages [23,24]. In two subsequent British Columbian cases (both occurring in the 1990s), the risks associated with language barriers to accurate diagnosis and informed consent were further illustrated. In the first of these cases, the British Colombia Supreme Court found a doctor ‘negligent in his examination and diagnosis of a man whose leg was amputated as a result of this misdiagnosis’ and awarded $1.3 million in damages ([25], p. 90). In its ruling, the Court stated that ‘the patient’s language difficulty should have made the doctor especially careful in conducting his physical examination’ ([25], p. 90).

In the second case, which involved a coronial inquiry into the death of a pregnant Vietnamese woman, language barriers were identified as a ‘contributing factor’ in the woman’s death [24]. The coroner in this case is reported to have recommended that ‘the College of Physicians and Surgeons assess the need for interpreters for patients who speak little or no English’ ([25], p. 90).

In another 1990s case, this time in the United States of America, a Mexican labourer (whose first language was Spanish) reportedly lost his sight as a result of a work-related eye injury not being properly assessed, diagnosed, and treated by an attending doctor at a clinic he attended for treatment. The labourer was struck in the eye by a 6 mm piece of metal while using a nail gun on a construction site. It was later revealed in the court case that followed that neither the doctor nor her assistant spoke Spanish and that the clinic did not provide an onsite interpreter. Although an interpreter was made available via a telephone interpreter service, the three parties involved (the doctor, the doctor’s assistant, and the patient) were not linked up via a speakerphone and the patient ‘never spoke directly to the interpreter’; only the doctor remained ‘on line’ during the consultation [26]. It was further revealed that the man: tried to communicate that he had been using a nail gun at the time of the accident and that a piece of metal struck his eye, However, it was noted in the clinic’s medical record that the patient had previously been hit in the eye with a wood chip. The clinic physician ultimately diagnosed the patient as having an abrasion to the eye and treated it accordingly [italics added] ([26], p. 1).

As a result of this miscommunication, the man’s eye condition was not properly diagnosed, which in turn resulted in a critical delay in appropriate treatment being administered that ultimately cost him his sight.

As well as cultural miscommunication, cultural misunderstandings and ‘cultural imposition’ have also been implicated in preventable adverse events in patients. A notable example of this can be found in the 1998 case of a 28-year-old Muslim woman who was awarded damages after her obstetrician and gynaecologist negligently sterilized her without her consent. The woman, who had originally come to Canada in 1991 as a refugee from Somalia, consulted with the doctor regarding a gynaecological infection that required a draining procedure. The woman did not speak, read, or write English and was accompanied by a friend who acted as an interpreter during the doctor’s appointment. The doctor subsequently performed a permanent sterilization procedure (notably, tubal ligation) on the woman, which she thought was a procedure to remedy her infection. Upon learning of the true nature of the procedure she had undergone, the woman brought legal action against the doctor for damages [27]. In reaching its decision, the court made explicit comment on the duty of a doctor to be ‘sensitive to cultural issues’—in particular, to ensure that, when practising in a multicultural society, doctors should take care not to impose their values and beliefs (inadvertently or otherwise) onto others who do not subscribe to them. The court’s comments in this matter are presented below: Ms Adan is a Muslim woman who had never used any form of birth control and whose reproductive capacity is fundamental to her status in society. Sterilization is not permitted under Islamic law. Although the observance of Islamic law can vary, I accept the plaintiff’s evidence that, despite Islamic law, she was never prepared to be sterilized. Dr Hinnawi [expert witness], who trained as a physician in Jordan and who has a large practice of Muslim patients, testified that he had never known a Muslim woman to undergo a sterilization. I do not suggest that Dr Davis knew or ought to have known anything about Islamic law or about the plaintiff’s particular religious practices. But, we live in a multicultural country where conformity to values and norms is variable and where careful inquiry must be made to ensure that our own values and norms are not inadvertently imposed on those who do not subscribe to them. It cannot be assumed that a 28 year-old woman with four children who are very close in age, and who has had a recent delivery by cesarean section, does not want any more babies. There was no careful inquiry here ([27], p. 278).

The court then went on to assert that not only do doctors have a positive duty to be culturally informed and to ensure that their patients fully understand information that is given to them, but also, in cases where interpreters are used, to be attentive to the linguistic ability of the interpreter and to ensure that patients whose first language is not the same as theirs are ‘returning reasonable and responsive replies’ ([27], p. 281). The court’s comments in this matter are presented below: Dr Davis was well aware of Ms Adan’s inability to speak English. In his evidence, he described her as ‘a patient we had difficulty communicating with’. That being so, it was incumbent on Dr Davis to make sure that he was understood which he failed to do. I do not minimize the challenges which physicians face in caring for patients who do not speak their language. I also accept that if consent is given through an interpreter who accompanies a patient, the physician must necessarily rely on the accuracy of the information which is imparted to the patient by the interpreter. But, the physician’s duty in these circumstances must surely extend beyond a standard recitation of a proposed surgery as if this were a patient who speaks the physician’s language and is of the physician’s culture. The physician must be attentive to the language ability of the interpreter. The physician must ensure that the patient is returning reasonable and responsive replies. If the patient is silent and asks no questions as was the case here, it is the responsibility of the physician to ask appropriate questions through the interpreters so as to be satisfied that the information has been understood. If Dr Davis had adopted this as his practice and if he had made a note which identified the interpreter and his understanding of the patient’s comprehension of the information, this could have been a very different case ([27], p. 281).

As well as demonstrating a link between culture, language, and patient safety outcomes, these above cases also demonstrate the risks associated with overestimating ‘interpreting’ per se, and underestimating the cultural variables that operate in a health interpreting situation and that may contribute to the incidence and impact of preventable adverse events during the clinical encounter. What is not always understood in regard to the issue of providing interpreting services to people of diverse cultural and language backgrounds is that intercultural communication requires not just an exchange of words (spoken sounds and conventional symbols) but also an exchange of shared meanings, which, in a clinical encounter, can be very difficult if not impossible when the people involved are from different racial, ethno-cultural, and linguistic backgrounds. Thus, to be effective, interpreters and bilingual staff must not only ‘translate words’ but establish and verify ‘shared meaning’. This requires not only knowledge of and skills in speaking a ‘second language’ but cultural knowledge and skills in an intercultural transactional approach to communication [28–30].

Lessons learned

It is important to clarify that none of the legal decisions cited here are binding on courts outside of their jurisdictions. Nonetheless, they contain important lessons for health care organizations and health care providers working in other jurisdictions. Firstly, they identify the standards expected of professional caregivers when caring for people of diverse cultural and linguistic backgrounds and who do not speak the same language as their professional care providers. Secondly, they warn of the importance of taking into account ‘cultural imperatives’ ([31], p. 6) in health care service delivery and of ensuring that patients understand the questions they are asked and the information that is given to them. This includes making sure that patients understand not just the words that are being exchanged but also the cultural meanings embedded in the words used. Thirdly, they underscore the responsibility of professional care givers to make ‘careful inquiry’ when caring for patients of different cultural backgrounds and to ensure that their own values and norms ‘are not inadvertently imposed on those who do not subscribe to them’ ([27], p. 278).

Finally, these cases demonstrate how failures to take into account pertinent cultural and language considerations in the clinical encounter may inadvertently expose patients of minority cultural and language backgrounds to what Reason [6] calls ‘a trajectory of accident opportunity’ for otherwise preventable adverse events while in hospital (e.g. wrong diagnoses, wrong site surgery, wrong procedures). They also demonstrate that unless cultural and language imperatives are addressed using a systems approach, they will translate into what James Reason calls ‘error-provoking conditions within the local workplace’ that, if not identified and managed in a proactive way, may ‘lie dormant within the system for many years before they combine with active failures and local triggers to create an accident opportunity’ ([6], p. 769).

Improving system processes

There is an obvious sense in which the stated role, aims, and related activities of various national patient safety organizations and hospital programmes are pertinent to ensuring safety and quality care of patients from minority cultural and language backgrounds. This, however, cannot be assumed without supporting evidence. In keeping with the tenets of Reason’s [6] model of human error management in health care, it is imperative that the vulnerabilities of patients from minority cultural and language backgrounds are identified and actively addressed via the development and deployment of systems and processes that would ‘make it hard for people to do the wrong thing and easy for people to do the right thing’ ([8], p. ix). To this end, the following strategies are recommended.

Firstly, comparative data on the incidence and impact of preventable adverse events in patients from minority cultural and language backgrounds compared with mainstream (majority) patient groups need to be collected. Differentiating data pertaining to incidents in which cultural and language considerations have been identified as significant contributing factors to a preventable adverse event also need to be collected. As Bagian et al. ([7], p. 524) have famously argued, ‘You can’t fix what you don’t know about’.

Secondly, attention needs to be given to examining systematically both organizational (system) and individual (person) processes contributing to preventable adverse events and ‘near misses’ in patients of minority cultural and language backgrounds. As in the case of medication errors, valuable learning opportunities could be gained by asking such questions as: What was the incident? Why did the incident occur? What was the proximal cause of the incident? What were the underlying system processes that possibly contributed to the incident? ([32], p. 40). Such questioning would also help to identify cultural communication and cultural practice variables (operating at either a system or person level) mediating preventable adverse events in culturally and linguistically diverse patient groups.

Thirdly, systematic processes need to be put in place that would make it difficult for an attending health professional not to engage an accredited health interpreter when doing a health assessment and prescribing needed cares and treatments for a patient whose language and cultural lifeways he or she does not understand. Such processes could include: employment of onsite health interpreters; ‘good’ referral and booking systems for cultural and language services; ready bedside availability of two-way telephone sets for telephone interpreting; employment of culturally competent bilingual health professionals; and cultural competency training of staff [10–13].

Finally, in order to develop evidence-based policy and practice initiatives in this area, a robust research agenda aimed at exploring and demonstrating the link between culture, language, and patient safety outcomes in minority cultural and language patient groups needs to be developed and operationalized at both a local and global level (see [33]).

Conclusion

All patients are entitled to expect and to receive safe care while in hospital. So long as patients of minority cultural and language backgrounds are at particular risk of experiencing preventable adverse events while in hospital, neither the health care system nor individual health care organizations can claim to be providing safe and quality care to patients. This is because a hospital or health service organization that is not identifying vulnerabilities in ‘the system’ and acting positively to either eliminate or reduce the vulnerabilities identified will not achieve its end goal of providing safe and better care to patients [7].

Patient safety policy and programme texts must contain explicit and substantive mention of the complexities and implications of the mediating variables of culture and language in the clinical encounter. Unless such explicit mention is made, the critical link between culture, language, and patient safety may be overlooked or minimized as questionable assumptions are made about the capacity of mainstream patient safety programmes to improve generally the safety and quality care of all patients regardless (as opposed to regardful) of their cultural and language backgrounds. The possible harmful outcomes of this oversight need to be anticipated and their risks reduced, if not prevented.

References

View Abstract