Patient neglect, defined as “the failure of a designated care giver to meet the needs of a dependent” [1] (p.437), has become an issue of concern in both North America and Europe [2, 3]. In the UK, this has been driven by media outlets [4, 5], charities [6], and health regulators [7]. Headlines such as “Want to know the NHS’s real problem? Ask a nurse for a bowl of cornflakes” [8], “Shamed hospital accused of leaving dying patients to starve” [9], and “Can patient neglect be a violation of human rights?” [10] capture concerns relating to patient neglect. They reflect public anxiety, with patients and families making 22,845 complaints to the NHS in 2011 on issues relating to staff attitudes, communication, and patient dignity [11]. Senior politicians acknowledge the issue, and argue that neglect has been “hidden away” [12] and that healthcare institutions must ensure “every patient is cared for with compassion and dignity” [13]. Solutions include “reducing stifling bureaucracy” [14], ensuring nursing staff talk to patients at least “once an hour” [13], utilising legislation and regulation to ensure staff consider patient’ “wellbeing and dignity” [15], and making staff sign-up to a “code of conduct” on dignity and respect [16]. The solutions reflect a belief that healthcare staff are responsible for instances of patient neglect, but they are also contradictory (e.g. reducing bureaucracy to free staff from form-filling whilst simultaneously increasing bureaucracy to ensure staff care for patients properly), or involve regulating aspects of behaviour that are difficult to measure and assumed to be lacking (e.g. compassion). These contradictions reveal the lack of a clear understanding of the nature and causes of patient neglect.

High-profile scandals have made patient neglect a key issue for policy makers. Scandals have included patients being regularly physically (e.g. left malnourished, dehydrated, in pain, and unwashed) or emotionally (e.g. being ignored whilst in need, not shown compassion, loss of dignity) neglected by healthcare staff [17–20]. Linking patient neglect to specific metrics of patient harm or clinical outcomes is difficult due to the often complex conditions of patients and their treatment [21]. Furthermore, conducting research is challenging due to the toxicity of the subject (e.g. questioning the abilities, motivation and ethics of staff) and a media narrative which seeks to blame rather than understand why poor care occurs [22, 23]. However, cases such as the Mid-Staffordshire NHS Foundation Trust scandal, where routine and basic failings in care resulted in up to 1,200 patients deaths between 2005 and 2008, show the catastrophic implications for patient care when neglect becomes systemic across an organisation [21, 24].

Researchers in medicine, health sciences, and psychology have for some time investigated how institutional processes, clinical environments, and the behaviour of healthcare staff influence patient safety [25]. These investigations have resulted in interventions (e.g. team-training, care bundles, skill validation) to reduce medical error and improve clinical outcomes [26]. Although they might be expected to reduce patient neglect, it appears necessary for practical (e.g. to meet public and political concerns) and conceptual reasons (e.g. to develop suitable interventions) to distinguish patient neglect from unintentional error, or intentional abuse. This is because reports on neglect such as those cited above often refer to: i) staff behaviours that may not directly lead to patient harm (e.g. not aiding patients to go to the toilet), but are crucial for care and probably do not reflect a competency gap; ii) staff attitudes and behaviours towards patients that cannot be regulated or easily measured (e.g. compassion); iii) a mixture of causal factors leading to patient neglect, some of which indicate neglect to be unintentional (e.g. due to a lack of resources) or alternatively not related to error (e.g. rudeness) [27]; iv) differing beliefs between patients, families, and staff as to whether neglect has occurred (e.g. for loss of patient dignity) and the causes of neglect; and v) breakdowns in institutional structures (e.g. communication between staff and management) that are a prerequisite to introducing interventions to improve care [26].

This is the first literature review on the nature and causes of patient neglect. Accordingly no protocol exists to guide the review, so standard protocols for literature review were applied [28]. The eligibility criteria were articles or reports published in English reporting primary data, since 1990, on the occurrence or causes of patient neglect anywhere in the world. In the first instance, the search for articles on patient neglect was framed using Lachs and Pillemer’s [1] (p.437), widely used definition (in reference to neglect of elderly patients) of “the failure of a designated care giver to meet the needs of a dependent”. From this perspective, patient neglect is behavioural (intentionally or unintentionally failing to meet the needs of a caregiver). The information sources, search terms used, and study selection procedure are outlined in Figure 1.

To evaluate the methodological quality of the research studies, we applied the SIGN system [29]. This provides ratings through which to assess the quality of data collected in quantitative and qualitative studies. The assessments for each study are reported in Table 1, with the quality ratings being the following:

1++: High quality meta-analyses, systematic reviews of RCTs, or RCTs with a very low risk of bias.

1+: Well-conducted meta-analyses, systematic reviews, or RCTs with a low risk of bias.

1-: Meta-analyses, systematic reviews, or RCTs with a high risk of bias.

2++: High quality systematic reviews of case control or cohort or studies. High quality case control or cohort studies with a very low risk of confounding or bias and a high probability that the relationship is causal.

2+: Well-conducted case control or cohort studies with a low risk of confounding or bias and a moderate probability that the relationship is causal.

2-: Case control or cohort studies with a high risk of confounding or bias and a significant risk that the relationship is not causal.

3: Non-analytic studies, e.g. case reports, case series.

The following data extraction exercise was performed. First, the meaning of neglect was reviewed in each paper. Second, behaviours identified in studies of patient neglect were identified. Third, frequencies of neglectful behaviours reported by healthcare staff, patients, and families were captured. Fourth, causal factors identified by articles and reports as contributing to instances of patient neglect were extracted. This included the capture of both qualitative data (TR) and quantitative data (AG). The extracted data were not amenable to meta-analysis due to a mixture of qualitative and quantitative studies being identified. Consistent with similar reviews of literature with mixed forms of data, a narrative analysis was used to synthesise the findings of the review [30, 31].

Figure 1 reports the results of the literature review. Ten research articles were included, with data largely collected in Scandinavia, South Africa, and the US. The majority of articles used survey methods to measure staff, family, or patient observations of neglectful behaviours [32–39]. Two qualitative papers investigated staff perceptions of patient neglect [40, 41], and patient perceptions of neglectful behaviours were also of interest [39, 42]. Several studies were conducted in elderly care. The hand search identified four qualitative UK government reports investigating patient neglect at both individual and unit/hospital level [24, 43–45]. Many discussion articles (e.g. on legal issues) and studies of related topics (e.g. patient dignity, ethics) were also identified, and were informative in understanding what is meant by patient neglect. However, they were not included in the review due a lack of relevant primary data focussing explicitly on patient neglect. The number of studies and reports seems to be increasing rapidly, with 8/14 (57%) being published between 2009–2012.

In comparison to the other literatures linking behaviours and outcomes in healthcare (e.g. medical error) [46], the number of studies investigating neglect is limited, and data was mostly descriptive. Quantitative investigations tended to be cross-sectional survey studies of staff and patients on how often patient neglect occurred in care institutions. These were relatively narrow in scope (e.g. observations of whether healthcare staff fail to wash patients), and did not provide outcome data. In terms of research quality, these studies generally scored 2- according to the SIGN guidelines [29], as their data were not causal and had quite a high probability of bias (e.g. likely social desirability effects in being questioned on neglectful behaviours). Nurses tended to report on instances of other staff showing neglectful behaviours, and patients and families on observations of patients being neglected. Although clinical outcomes of patient neglect were cited as bedsores, malnutrition, infections, dehydration, contractures, and early mortality, no study systematically linked them to behaviour [37]. Furthermore, causal factors underlying perceptions of neglect were not statistically associated, and were discussed in the generality. For example, failures to feed patients were repeatedly cited as being examples of patient neglect, but it was not clear as to whether these occurred due to external system failures (e.g. catering problems), workloads (e.g. too many patients to manage), crises (e.g. diverting attention), team failures (e.g. confusions of responsibility), unintentional error (e.g. incorrect beliefs on patient diet), or a lack of a caring attitude towards patients.

In the qualitative studies and four government reports, specific patient encounters and failings in care were described. In terms of research quality, these studies generally scored 3 according to the SIGN guidelines [29], as their data were derived from case reports and interviews with staff, patients, and families. These often described extreme neglect (e.g. malnutrition contributing to patient mortality), and tended to focus on clear procedural breaches by staff (e.g. pain management protocols). They also focussed on the perspectives of patients and families on why neglect occurred (and feelings of not being cared for), however these were often general and not linked to specific behaviours. As indicated above, there was a lack of clarity for what the research publications meant by ‘patient neglect’. The following section considers further the meaning of patient neglect in order to fully distinguish it from concepts such as error, and to reflect its apparent subjectivity.

There is growing public concern over patient neglect in healthcare institutions. To understand and explain what patient neglect is, and why it occurs, it has been necessary to draw on a range of psychology literatures. These include the error and workplace deviance literatures [51, 63, 65], the patient safety, organisational psychology, and non-technical skills literatures [23, 27, 101], and social psychological research on perspective taking, burnout, and caring [55, 59, 61, 68]. Using these literatures, we have conceptualized patient neglect as clearly distinct from error (purposeful actions with unintentional consequences) and abuse (purposeful attempts to inflict physical or emotional harm). In order to contribute to public dialogue and academic analysis, we have differentiated between caring and procedure neglect. Whilst procedure neglect refers to the objective measures of patient neglect (e.g. not feeding patients) that might result from error, abuse, or lack of caring, caring neglect attempts to capture the subjective and emotional aspects of poor care. Caring neglect can be damaging to the emotional well-being of patients (e.g. dignity) even if subjective beliefs about staff attitudes are not accurate. The concept of caring neglect appears important for addressing public concerns because this is often what patients and their families are worried about, yet it is largely invisible to healthcare institutes because it refers to behaviours which have not been (and possibly cannot be) fully proceduralised. Central to caring neglect are the interactions, perceptions, and beliefs of healthcare staff and patients. It relates to behaviours that are physically or emotionally neglectful towards patients, represent a perceived or actual lack of caring and compassion for patients.

Consistent with literatures on accident causation [25, 102], a range of proximal factors (e.g. workloads) were found to shape the medical and nursing care provided to patients, and these were in turn influenced by distal (e.g. institutional) factors such as organisational leadership. The review data does not enable us to separate the causal pathways leading to caring neglect from the pathway leading to either error or abuse. However, it does seem that caring neglect can lead to a distinctive negative outcome, namely, emotional patient harm from loss of dignity or feeling uncared for. Crucially for developing interventions, patient neglect is found to occur due to both factors relating to staff attitudes, and others factors relating to such as poor communication training, or systems (overlapping with the medical error literature). Figure 2 presents an initial psychological model of patient neglect. It illustrates the findings of the review to explain why neglect occurs, and considers the potential outcomes of neglect. These require further study, and future modelling should consider the role of error in understanding why neglect occurs.

Only ten research articles were identified as reporting primary data on patient neglect in healthcare settings. Few academic articles were published in the UK, despite the public concern. This may reflect i) a dearth of research into the subject, ii) the use of inappropriate search terms, or iii) that research investigating patient neglect is obscured within other studies focussing upon error or abuse. However, it is notable that in conceptualising patient neglect, our review referred informally to other key health service research literatures. For example, it considered medical error and patient abuse, both of which can describe behaviours associated with procedure neglect, albeit in the context of mistakes or intentional harm. It also referred to work investigating what patients consider to be ‘good care’. This often relates to interactions between healthcare staff and patients (good technical care is assumed) [56]. Professional, approachable, and compassionate attitudes towards patients, as embodied by caring behaviours, are seen as the bedrock of good care [55, 56, 103]. Their absence is an element of caring neglect [104], and although research on good care does not focus explicitly on the neglect of patients, it may aid our understanding of patient neglect.

However, current methodologies for studying error (e.g. retrospective record review) or good care (e.g. audits) may struggle to assess why patient neglect occurs, and new research techniques are required (e.g. focussing on complaint data). This is perhaps demonstrated in the ratings of research quality for the publications included in the review [29] – studies were largely assessed as being descriptive and non-predictive. Further work is required to understand the causes of both procedure and caring neglect, and this analysis was restricted to publications in the literature review. Other factors may also be important, including clinical training, individual differences (personality, motivation), socio-economic factors, and expectations on care. Furthermore, the international nature of the research on patient neglect demonstrates wider interest in the topic. This is in turn poses questions about patient neglect in countries with different institutional procedures and/or different patient expectations, and consequently the distinctive steps that might be taken to improve care.

In this article we developed a social psychology-based conceptual model to explain the occurrence and nature of patient neglect. To date, patient neglect has been difficult to conceptualize because it has one component which is defined by institutional procedures (procedure neglect) and another component which is defined by patient expectations about care (caring neglect). Separating these components reveals a possible divergence of perspective between medical institutions and the public on the nature of patient neglect. Moreover, it leads to new questions: Can the behaviours which lead patients to feel neglected become proceduralised or measured? And if not, then how can one ensure that staff engage in the multitude of non-measurable behaviours which lead patients to feel cared for? This is a critical question for patient safety specialists and policy makers, as it indicates regulations and targets are unlikely to reduce the caring neglect that is often the focus of patient complaints [11]. Indeed, creating rules that reduce the ability of healthcare staff to engage in discretionary caring acts (e.g. filling forms on caring) may aggravate the problem. To begin to address these questions, future research must i) further investigate healthcare staff and patient perceptions of patient neglect, ii) systematically investigate the relationship between caring neglect and procedure neglect and their impact upon health outcomes, and iii) consider the appropriateness of current patient safety interventions for reducing patient neglect.