A knowledge, attitude, and practice survey on mediation among clinicians in a tertiary-care hospital in Singapore

Healthcare delivery is a highly complex, deeply personal and costly endeavour that involves multiple specialties and services. There is an imbalance in knowledge between the healthcare provider and consumer that may contribute to doubts and uncertainty over treatment and outcomes. It is unsurprising that conflict and dispute can develop between healthcare providers and patients and their next-of-kin. The use of mediation in the healthcare setting has recently been promoted in many developed countries, including Singapore. We administered a detailed 32-item survey in a large tertiary-care teaching hospital to improve our understanding of the knowledge, attitude and practice of dispute resolution among clinicians to pave the way for better strategies to improve the adoption of mediation in healthcare setting. Ninety-seven respondents had an average of 62% (SD: 12%) knowledge score. The most common misconceptions held by the respondents about mediation were: (1) mediation was about fact-finding, (2) mediation is limited to only certain types of dispute, (3) mediation proceeds by both parties giving their account of the dispute, then a third party decides a settlement, (4) the average time it takes to resolve a dispute through mediation, (5) the cost of mediation, (5) the venue of mediation, (6) the person determining the outcome of mediation, (7) confidentiality of mediation. In general, the respondents were positive about the use of mediation as a dispute resolution tool. When asked to indicate the relative importance of different outcomes of dispute resolution, financial compensation and waiver of hospital bill attracted mixed responses while understanding facts of dispute, assurance that the same error would not recur, and offering corrective treatment were rated as being important. By contrast, seeking an apology from the complainant was considered neutral to somewhat important and the respondents were least concerned with the publicity of the dispute. Direct negotiation with the complainant was considered the most time- and cost-efficient means of resolving a dispute while the opposite was true for litigation. Mediation was considered the approach where the clinicians are most likely to achieve their desired outcome while litigation was considered least likely to produce a favourable outcome. Approximately half of the respondents reported having personal experience or known of a colleague who had been involved in a medico-legal dispute. A quarter of these cases were resolved by direct negotiations with the complainant while lawyers, the judge and mediation, resolved approximately 15% each, respectively. The knowledge base of the clinicians in this study about mediation was moderate and probably reflected the general lack of direct experience in the resolution of a dispute or training in mediation. This further corroborated with the general response that the uptake of mediation in the healthcare setting is currently poor in Singapore due to the lack of awareness and perceived lack of avenue among the surveyed clinicians. Any further work to be done with clinicians may be in the direction of (1) increasing general understanding of mediation, (2) increasing awareness of avenues for mediation, and (3) becoming better aware of when to propose mediation.


Introduction
Healthcare delivery is a highly complex and costly endeavour that involves multiple specialties and services. It is also deeply personal and often evokes strong emotions not only in the patient, but also the caregivers. Moreover, there is an imbalance in knowledge between the healthcare provider and consumer that may contribute to doubts and uncertainty over treatment and outcomes [1,2]. It is unsurprising that conflict and dispute can develop between healthcare providers and patients and their caregivers.
There is an increasing number of disputes in the healthcare setting being sent for litigation [3,4]. More recently, open communication has been advocated as a means to reduce the occurrence of dispute in the healthcare setting [5]. Open communication is a practice of honest disclosure to the patient and caregivers when things do not happen as expected. Open communication can resolve a dispute before it reaches the litigation stage. Nevertheless, there are still disputes that cannot be fully resolved by open communications. For these cases, alternate dispute resolution strategies have been gaining traction. Among the alternate dispute resolution strategies, mediation is an important tool to achieve an amicable settlement that is acceptable between the parties [6][7][8].
The use of mediation in the healthcare setting has recently been promoted in many developed countries, including Singapore. We undertook a detailed survey in a large tertiary-care teaching hospital to improve our understanding of the knowledge, attitude and practice of dispute resolution among clinicians to pave the way for better strategies to improve the adoption of mediation in healthcare setting.

Materials and methods
The National University Hospital, Singapore is a 1100-bed tertiary-care teaching public hospital serving the Western half of Singapore. It is affiliated to the National University of Singapore and provides a complete suite of medical and surgical services that encompasses obstetrics and gynaecology, paediatrics surgery and medicine, adult surgery and medicine, and orthopaedics. We received ethics approval for this study (National Healthcare Group Domain Specific Review Board, reference number: 2016/01362).
We developed a 32-item questionnaire that probed the knowledge, attitude and practice (based on past experience as well as hypothetical scenario) of mediation and dispute resolution of the clinicians through a mix of open-and close-ended questions ( Table 1). The respondents were also asked to indicate their preferences using Likert-scale and ranking. The complete questionnaire is provided alongside the results in the following section. The survey questionnaire was developed by the investigators and piloted on a small group of 7 respondents to refine and improve it. Once finalised, the questionnaire was adopted on the Google survey platform. An email invitation was sent to all the clinicians in our institution (n = 1264) with a reminder email sent one and three months later. Additionally, the invitation to participate in the questionnaire survey was placed on screen before the weekly grand round in our hospital for four months. The survey participants provided informed consent electronically prior to undertaking the questionnaire survey.
The respondents voluntarily participated in the survey and provided implied consent to have their data analysed in an anonymised, aggregated manner. The survey did not collect personal or identifiable data.

Statistical analysis
The survey responses for each question were summarised using simple descriptive statistics. The percentage of the response was calculated with the total number of participants (i.e. n = 97) as the denominator. Subsequently, the participants were then divided into those who had prior experience in managing conflict and those who had not. The responses to the knowledge and attitude questions (i.e. Questions 1 to 17) between the two groups of participants were compared using Fisher's exact test (two-tailed).
The knowledge of the participants was also scored against idealised/preferred answers for Questions 1 to 13, which were determined by the investigators. For questions with only one 'model' answer, a matched response was given a score of 1 while any other response was scored 0. For questions with multiple preferred answers, the respondents were scored 1 if they have fully matched responses, 0.5 if they had partially matched responses and 0 if they have fewer than half matched responses. The scores were then expressed as a percentage of the maximum of 13. The knowledge scores were correlated with the preferences of the clinicians for Questions 14, 15 and 20 in the Attitude section. A p-value of <0.05 was considered significant in this study.

Results
In total, we received 97 responses from the clinicians, representing a response rate of 8.7% from 1121 eligible clinicians in our hospital. Nearly two-thirds of the respondents were senior staff who had more than 6 years of working experience (65/97, 67%), followed by junior staff who had 0-3 years of working experience (19/97, 20%) and those with 4-6 years of experience (13/97, 13%). Correspondingly, 62 (64%) clinicians had postgraduate qualifications. Just less than half (45/97, 46%) of the clinicians were female.
The responses to the individual questions are summarised in Table 1 below. The preferred/ idealised responses for the knowledge-related questions are highlighted in italic.

Knowledge of the respondents
The respondents had an average of 62% (SD: 12%) knowledge score. The most common misconceptions held by the respondents about mediation were (see Questions 1 to 13): (1) mediation was about fact-finding, (2) mediation is limited to only certain types of dispute, (3) mediation proceeds by both parties giving their account of the dispute, then a third party decides a settlement, (4) the average time it takes to resolve a dispute through mediation, (5) the cost of mediation, (5) the venue of mediation, (6) the person determining the outcome of mediation, (7) confidentiality of mediation.

Attitude of the respondents
In general, the respondents were positive about the use of mediation as a dispute resolution tool (Questions 14 and 15). The knowledge scores did not correlate with the inclination of the clinicians to learn or use mediation as a dispute resolution tool (correlation coefficient <0. 3). An overwhelming majority (94%) of respondents felt that maintaining a good relationship with the complainant was important (Question 16a). When asked to indicate the relative importance of different outcomes of dispute resolution (Question 16b), financial compensation and waiver of hospital bill attracted mixed responses while understanding facts of dispute, assurance that the same error would not recur, and offering corrective treatment were rated as being important. By contrast, seeking an apology from the complainant was considered neutral to somewhat important and the respondents were least concerned with the publicity of the dispute.  Direct negotiation with the complainant was considered the most time-and cost-efficient means of resolving a dispute while the opposite was true for litigation (Questions 17-19). Mediation was considered the approach where the clinicians are most likely to achieve their desired outcome of dispute resolution (Question 17) while litigation was considered least likely to produce a favourable outcome.

Practice of the respondents
Just over half of the respondents (54%) had experience managing a dispute with patients (Question 22). Of these, half had managed five or more disputes in the past. Interestingly, prior experience in managing dispute did not produce statistically different responses to the knowledge-and attitude-related questions (Questions 1 to 17). Nevertheless, the overall knowledge score was higher in respondents with experience managing a dispute with patients compared to those without such experience (65% vs 58%, p <0.005). Approximately half of the respondents reported having personal experience or known of a colleague who had been involved in medico-legal dispute (Question 25). A quarter of these cases were resolved by direct negotiations with the complainant while lawyers, the judge and mediation, resolved approximately 15% each, respectively.
When the clinicians were asked to put themselves in the shoes of the patients (or their next of kin), they mostly preferred to raise concerns about their care with a fellow clinician (Question 27). The clinicians would choose not to raise a negative experience with a hospital when they consider it as a minor issue (61%) or because they wish to maintain a good relationship with the colleague/ hospital (36%)(Question 28). They were somewhat likely to resolve the matter through a neutral third party or raising it to the hospital authority than considering legal advice (Question 29).

Discussion
Traditionally, public healthcare institutes have in-build mechanisms to manage disputes. The majority of cases are handled by such quality service management mechanism, the "direct negotiation" route, where a hospital representative will investigate the complaints and respond to the patient or next-of-kin appropriately. Direct negotiation is generally the preferred tool that a healthcare institution uses to resolve dispute since it is important for the parties to establish direct communication to ascertain the details of the dispute. This tool is particularly useful if the communication/ content is straightforward, and there is no ill will between the parties and parties are clear about their positions. However, in the event of escalation, the dispute will proceed in the direction of litigation, where the insurers and lawyers of both the hospital and the clinician, will be activated to defend against the claims.
Mediation is a voluntary and confidential conflict resolution process where a neutral third party, the mediator, works alongside the parties to find mutually agreeable solutions to the dispute at hand. It is useful when the relationship between the parties is strained and direct communication has reached an impasse. It is also useful if one party is seeking non-monetary resolution (such as apology, acknowledgement, change in process), which the litigation process cannot provide for. The mediation process caters a confidential platform for parties to explain, exchange and clarify information, restore relationship/ trust. These are elements to an amicable settlement, or sets parties on the path to reach a sustainable and enforceable settlement. There are various accreditation schemes in Singapore to train and accredit mediators. Such training equips mediators with a framework and skillsets to facilitate discussion between disputing parties to achieve better communication, appreciate different perceptions, manage emotions, and overcome impasses. The mediators do not make any orders or decisions on the outcome.
The mediation movement in Singapore became relatively more active in the 1990s and since then the courts actively promote and encourage the use of mediation for suitable disputes [9]. The mediation movement for the healthcare industry came into spotlight in recent years. With increasing insurance premiums, a potential upward trend of medical litigation and more complex care required due to the aging population, the Ministry of Health Holdings set up its Mediation Unit in 2014 and re-introduced the Healthcare Mediation Scheme to encourage take up on mediation for healthcare-related disputes. Prior to that, the Singapore Medical Council (SMC) and the Singapore Mediation Centre piloted a SMC-Medical Council Mediation Scheme for SMC Complaints Committee to refer matters for mediation (Section 42(4)(b) (ii) of the Medical Registration Act). According to the Head of Mediation Unit, we note that to-date, Healthcare Mediation Scheme has since facilitated the mediation of 25 cases, of which 18 cases were settled, 6 were not, and 1 of them is pending negotiations. The settlement rate is approximately 75% (private communication by the Healthcare Mediation Unit at Ministry of Health Holdings Pte Ltd).
The Honourable the Chief Justice, Sundaresh Menon, highlighted in his speech of 28 October 2014 to the Obstetrical & Gynaecological Society of Singapore that (1) the tort of negligence and the adversarial nature of the litigation system do not provide a holistic solution for disputes relating to bad medical outcome; (2) an increase in medical malpractice law suits will lead to more clinicians practising defensive medicine and result in a lose-lose situation for both individuals and clinicians.
Identifying the main issues of disputes in healthcare industry as "dashed expectations and miscommunications", mediation, which is a process that focuses on improving communication, active listening and extra-judicial resolutions, was proposed to be the first port of call for potential medical negligence disputes [9].
Some pertinent observations from this survey: 1. There is an expectation of mediation to go beyond 7 days (Question 5) and yet a large group shows willingness to use mediation to resolve dispute (Question 15).

2.
Despite not having training in mediation, the majority of the surveyed participants showed above average knowledge of mediation.
3. Mediation may be a suitable platform for clinicians to explain the circumstances and treatment to the patients or their next-of-kind, as compared to litigation. This is because the clinicians surveyed had indicated that "understanding the facts of the dispute" as an important outcome of mediation. Aside from mediation, direct negotiation could also be explored as achieving this outcome, albeit without the presence of a moderating third party (mediator).
4. Most clinicians find that the litigious route is the least likely to obtain desired outcome and will be most time consuming and resource intensive. While the majority perceived direct negotiation to be the least time consuming and least expensive.
5. The study shows that the majority of respondents are keen to use mediation in the event of a conflict and are sufficiently interested to learn about mediation, even though mediation may not be the most resource efficient option to them.
6. The majority has no training in relation to management of conflict/dispute in the hospital setting (Question 23a) and are keen to learn more about it.
Based on the above observations, clinicians do see the advantages of mediation and that their desired outcome cannot be achieved by litigation if there is a material dispute with patients or their next-of-kin. Any further work to be done with clinicians may be in the direction of (1) increasing general understanding of mediation, (2) increasing awareness of avenues for mediation, and (3) become better aware of when to propose mediation. It is hoped that an increase in educational effort in these areas will improve the uptake of this dispute resolution tool. Nevertheless, the need for continued professional training in clinical care may reduce the priority in the learning of a non-medical related topic. The challenge therefore is how to build the mindset that learning to prevent or resolve disagreement is a key component of holistic patient care, as important as learning about the effects of the latest drug.
This study provided helpful insights into the knowledge, attitude and practice of mediation and dispute resolution in a public healthcare institution in Singapore. The knowledge base of the clinicians in this study about mediation was moderate and probably reflected the general lack of direct experience in the resolution of a dispute or training in mediation. This further corroborated with the general response that the uptake of mediation in the healthcare setting is currently poor in Singapore due to the lack of awareness and perceived lack of avenue among the surveyed clinicians. To help equip future generation of doctors with the necessary skills to navigate an increasingly litigious healthcare environment, medical students should be introduced to mediation and negotiation skills that have been tailored to the healthcare setting and introduced through existing modules relating to patient communication.
There are several limitations to this study. Firstly, this study was conducted in only one tertiary-care teaching hospital and the findings may not be easily generalisable to other clinical settings (e.g. primary-care). Secondly, the response rate from the clinicians is low, which may increase the risk of non-response bias. This may skew the overall responses towards respondents who already have an interest in mediation and have better knowledge about it. Consequently, the results obtained from this survey should be interpreted with care and should be seen as a 'best case' scenario. Nevertheless, the spread of the experience respondents is representative of the overall clinician population in our hospital and should provide a reasonably representative view of the clinicians in our institution.