1 The respondent operates an international school in Singapore. The claimant had been employed as an administrative assistant in its music department since 2008. Then, in July 2024, she was diagnosed with stage 3 nasopharyngeal cancer and underwent months of chemotherapy and radiotherapy. When that ordeal was finally behind her, her oncologist certified her fit to return to work on 26 June 2025.

2 But she did not return as she had left. She came back slower, weaker in her hands, needing rest and rehabilitation. She asked for accommodations to assist her recovery and her transition back to work. The respondent, she felt, was unwilling to meet her where she was. She said that this response took a toll on her health. Within six months of her return, she resigned, and her career of 17 years with the respondent ended like this.

3 She alleged that the respondent’s unwillingness to reasonably accommodate her medical needs amounted to breaches of two implied terms of her contract of employment—the duty to take reasonable care of the health and safety of employees, and the duty of mutual trust and confidence. These breaches, she contended, left her with no real choice but to resign, such that she was properly to be regarded as having been dismissed for the purposes of her claim under s 14(2) of the Employment Act 1968 (2020 Rev Ed) (the “EA”).

4 It is worth pausing at the outset to situate these implied terms within the broader architecture of the employment contract. Employers, as a general matter, hold the right to direct the labour of their employees as they see fit. This managerial prerogative finds expression through terms implied by law: employees are required to obey the lawful instructions of their employers (see Lister v Romford Ice and Cold Storage Co Ltd [1957] AC 555 at 594), and to serve their employers faithfully within the bounds of the contract (see Ticehurst v British Telecommunications [1992] IRLR 220 at [52]).

5 While “[t]hese implied terms help a contract of employment to function efficiently”—sparing the parties the impossible task of specifying in advance, with any precision, every task an employee might be called upon to perform—they simultaneously “leave an employee in a subordinate and vulnerable position” vis-à-vis her or his employer: see Hugh Collins, “Implied Terms in the Contract of Employment” in The Contract of Employment (Mark Freedland et al eds) (Oxford University Press, 2016) ch 22 (“Collins, ‘Implied Terms in the Contract of Employment’”) at 478. To address this vulnerability within the structure of employment, the law implies, in turn, countervailing obligations upon the employer. These include the duty to take reasonable care of the health and safety of employees, and the duty of mutual trust and confidence.

6 The existence of the implied term requiring employers to take reasonable care of the health and safety of employees is well established: see Johnstone v Bloomsbury Health Authority [1991] IRLR 118 (“Johnstone”). I note that analogous obligations also arise under statute (see s 12(1) of the Workplace Safety and Health Act 2006 (2020 Rev Ed)) and under the general law (see Chandran a/l Subbiah v Dockers Marine Pte Ltd [2010] 1 SLR 786 at [13]).

7 The implied term of mutual trust and confidence stands on somewhat less settled ground, at least in Singapore. The Appellate Division of the High Court observed in Dong Wei v Shell Eastern Trading (Pte) Ltd [2022] 1 SLR 1318 that its status “has not been clearly settled in Singapore” (at [82]), a remark that echoed the Court of Appeal’s earlier observation in The One Suites Pte Ltd v Pacific Motor Credit (Pte) Ltd [2015] 3 SLR 695 (at [44]) that the position in Singapore on this implied term was “still left open for decision in a future case”. That said, the term has been recognised and applied at the High Court level on several occasions, most recently in Prashant Mudgal v SAP Asia Pte Ltd [2026] 3 SLR 914 (“Prashant”). As the General Division of the High Court (“GD”) observed in that case (at [122]), such precedent establishes that “the implied term has existed and continues to exist in employment contracts under Singapore law” unless and until it is expressly overruled.

8 The GD in Prashant went further, opining that considerations of fairness and policy militate in favour of implying the duty of mutual trust and confidence (at [137]). An employment contract, the GD noted, “involves a longer-term relationship between the parties in which they make a substantial commitment” (at [138]). The power imbalance inherent in that relationship, “coupled with the paramount role which a person’s occupation plays in his sense of identity and self-worth, makes employees especially vulnerable vis-à-vis their employers” (at [140]). The GD also endorsed (at [142]) Lord Steyn’s observations in Johnson v Unisys Ltd [2003] 1 AC 518 (at [19]) that with “the greater pressures on employees due to the progressive deregulation of the labour market, the privatisation of public services, and the globalisation of product and financial markets… [t]he need for protection of employees through their contractual rights, express and implied by law, is markedly greater than in the past”.

9 These observations bring us full circle to Professor Collins’s insight with which we began (see [5] above): that the structure of employment places employees in a subordinate and vulnerable position, and that the law—whether through statute or judicial decision—responds by imposing countervailing duties upon the employer. As Professor Collins puts it (see Collins, “Implied Terms in the Contract of Employment” at 479-80):

10 Hence, while the employer’s prerogative to direct labour is incontrovertible, it exists in relationship with the employer’s corresponding responsibility to the employee to exercise it with care, and without destroying the trust and confidence upon which the employment relationship depends. The claimant did not challenge the respondent’s right to direct her labour. What she challenged was the way that right was exercised in the face of her medical vulnerability—and whether, in so exercising, the respondent fell short of its own implied obligations.

11 Before proceeding, a word about the structure of this judgment is in order. The material facts concern the six-month period following the claimant’s return to work on 26 June 2025; her last day of employment was 18 January 2026. They are largely undisputed, and I shall note where genuine disputes of fact arose. I shall examine the matter in a broadly chronological manner through the material six-month period, weaving the legal analysis into the narrative as each episode demands. The implied duty to take reasonable care of the health and safety of employees provides the primary lens through which the facts are examined. The implied term of mutual trust and confidence, which draws upon the cumulative picture, is addressed thereafter. I then turn to address the question of whether the respondent’s course of conduct had forced the claimant to resign, such that she was to be treated as having been dismissed within the definition under s 2(1) of the EA, and whether the claimant had been dismissed without just cause or excuse.

12 After enduring months of chemotherapy and radiotherapy, the claimant’s cancer was in remission. Her oncologist, Dr MO, in a memorandum dated 19 June 2025 (the “First Oncologist Memorandum”), assessed her as fit to return to work on 26 June 2025. Remission, however, does not mean full recovery; fitness to return to work does not mean fitness to work as one did before. It is common knowledge that chemotherapy and radiotherapy can leave lasting side effects on those who undergo them. In the claimant’s case, those side effects manifested as joint pains, reduced strength, and lingering fatigue. She was fit to return to work in the sense that her primary functions—providing administrative and secretarial support for the efficient day-to-day operation of the music department—did not demand physical exertion beyond what she could manage. But the side effects meant that she would experience a degree of physical strain and exhaustion that prevented her from returning to her previous levels of productivity. Therefore, Dr MO added in the First Oncologist Memorandum an important qualification: “I recommend light duty work for [the claimant] since she is suffering from side effects to treatment”.

13 Moreover, cancer survivors in the claimant’s position typically receive structured support through a return-to-work programme at the Singapore Cancer Society (“SCS”) rehabilitation clinic, which provides—as I understand it—free rehabilitation sessions to assist reintegration into the workforce. The claimant had been attending twice-weekly sessions there since April 2025. She was also required, on her own part, to maintain a daily programme of exercises and stretches prescribed by the clinic to support her ongoing recovery.

14 A senior occupational therapist at the SCS rehabilitation clinic, Ms OT, had written a memorandum for the claimant on 19 June 2025, explaining her participation in the programme and noting that she was experiencing “ongoing side effects from her cancer journey that might affect her return to work”. Ms OT further recommended that the claimant avoid handwriting where typing was possible, refrain from carrying heavy items, and take short breaks during the day to manage her fatigue—practical accommodations that asked relatively little of any employer in respect of an employee whose primary functions were administrative in nature.

15 The claimant furnished the respondent with the First Oncologist Memorandum as evidence of her fitness to return. She duly returned on 26 June 2025, and her transition back to work began promisingly enough. It was then the school’s summer vacation; there were no students on campus and no educational programmes running. The claimant and her colleagues worked from home during that period. That arrangement proved well-suited to her circumstances—the flexibility it afforded allowed her to attend her twice-weekly rehabilitation sessions, keep up with her daily exercises and stretches, and take the breaks her condition required.

16 On 31 July, an HR executive wrote to the claimant by email to clarify whether the light duty recommendation in the First Oncologist Memorandum was subject to a fixed period, whether any specific requirements or restrictions attached to it, and whether Mr LM—her line manager and the head of the music department—had been informed. The claimant replied the following day. She explained that Dr MO was “not able to give a fixed light duty period” given that her recovery would vary “based on individual response,” but that her condition would be reviewed periodically and updates provided. She also furnished Ms OT’s memorandum and confirmed that she would speak to Mr LM once he was back on campus.

17 At this point, the respondent had in its possession two memoranda representing the considered views of two healthcare professionals. Where Dr MO had left the nature of the claimant’s vulnerabilities and medical needs somewhat unspecified, Ms OT supplied the particulars: the joint pains that made handwriting difficult and carrying heavy items impossible, the fatigue that made sustained effort without rest a burden, and the ongoing rehabilitation sessions that were an integral part of her recovery.

18 An ordinarily prudent employer reading both memoranda together would have comprehended that the claimant’s return required careful handling, and that its duty to take reasonable care of her health and safety carried heightened demands compared to those owed to employees at large. That duty is not uniform in its practical content—it varies with the circumstances of the individual employee. Paris v Stepney Borough Council [1951] AC 367 (“Paris”) illustrates the point well. Though that case concerned the analogous duty arising in tort, it remains instructive for the contractual duty as well: see Johnstone at [20]. In Paris, the majority of the House of Lords held that an employer owed a heightened duty of care towards a one-eyed employee, having regard to the grave consequences that an injury to his remaining eye would entail. As Lord Oaksey put it (at 384):

19 Likewise, it was plain that the claimant’s circumstances were far from ordinary. She was still in active recovery, medically vulnerable after intensive cancer treatment, and the side effects of that treatment continued to bear upon her capacity to work. Discharging the duty owed to her could not look the same as discharging the duty owed to the respondent’s employees generally.

20 What, then, would the practical content of the duty to take reasonable care look like in relation to the claimant? This was not a case where the required steps to be taken by the employer were self-evident—as in Paris, where the risk of injury to the eye plainly called for protective eyewear, or as where an employee laid low by illness plainly requires rest. The side effects of cancer treatment—the fatigue, the joint pain, the physical fragility—tend not to manifest themselves visibly to those around the sufferer, and it is not always obvious what a recovering cancer patient needs in order to return safely to working life. And although the memoranda of Dr MO and Ms OT made clear that the claimant was suffering from the side effects of treatment and required light duty accommodations, medical documentation of that kind does not specify with any precision how a particular employee should or should not work. Every job is different, and it would be unrealistic to expect healthcare professionals to descend to that level of operational granularity on behalf of their patients.

21 There was a further consideration. At the time the respondent received those memoranda, it was still the summer vacation. The demands on the claimant were correspondingly light—there were no students on campus, no educational programmes running, and the general pace of the school’s operations was subdued. That would change once the school term resumed. Students, teachers, and parents would return, and with them the full weight of the administrative and secretarial demands that the claimant’s role entailed. The practical burden on the claimant would increase in step with the school’s operational needs. The respondent, of course, retained its prerogative to direct her labour in accordance with those needs. But that prerogative was not without limit. It could not be exercised in a manner that placed the claimant’s health and safety at risk, and the respondent could not shelter behind managerial authority to excuse a failure to discharge its implied obligations towards her.

22 How, then, was an ordinarily prudent employer—one with notice of the claimant’s vulnerabilities and medical needs—to navigate this situation, particularly once the school term resumed? The observations of the United Kingdom’s Employment Appeal Tribunal (the “EAT”) in British Aircraft Corporation Ltd v Austin [1978] IRLR 332 (“Austin”) are instructive on this point. The EAT stated (at [17]):

23 Austin concerned an employee who alleged that her employer had breached its duty to take reasonable care of her health and safety by failing to properly investigate her complaint that no suitable eye protection was available for someone who, like her, wore spectacles. The nature of her work required eye protection, but she found that the protection provided was incompatible with her prescription lenses and sought safety glasses that could accommodate them. Her employer did nothing—it did not investigate the matter, and nothing was done.

24 On those facts, the EAT affirmed the first-instance tribunal’s finding that the employer had breached its obligation to take reasonable care by failing to act reasonably in response to a safety concern drawn to its attention. The EAT did not prescribe what a proper investigation ought to have concluded—whether, for instance, the employee could have managed with standard protection or required bespoke eyewear. The breach lay not in any particular outcome but in the failure to engage with the complaint at all. As the EAT put it (at [11]):

25 The principle that emerges from Austin is that the duty to take reasonable care requires, as a starting point, that the employer reasonably investigate, consider, and respond to matters of health and safety drawn to its attention. Applied to the present case, before even addressing the question of whether the respondent ought to have granted the claimant any specific accommodation, there is the antecedent question of whether it even took her medical situation seriously and responded to it with the care that an ordinarily prudent employer would have brought to bear.

26 The school term resumed on 7 August 2025, and the claimant returned to campus that day. On that first day back, she spoke to Mr LM—who was also her line manager—about her medical status and her needs. She told him about her fatigue, her restrictions on carrying heavy items, her need for short breaks, and her twice-weekly rehabilitation sessions. Mr LM indicated that he was content with her taking short breaks as she needed and attending her twice-weekly sessions.

27 The claimant also asked whether she could work from home on an additional day each week. She had been working from home on Fridays prior to her cancer diagnosis, and she sought one further day as a temporary accommodation during her recovery. According to the claimant, Mr LM agreed, and she accordingly began working from home on two days each week—Fridays as a fixed day, and a second day to be determined depending on how she felt physically.

28 The respondent did not challenge the claimant’s evidence that Mr LM had essentially acquiesced to this arrangement. It did, however, provide context. The claimant had originally begun working from home on Fridays as a result of safe distancing measures during the COVID-19 pandemic, when all employees were permitted to work from home on specified days. Once those measures eased and school activities resumed at pre-pandemic levels, the working-from-home arrangements for educational support staff—of which the claimant was one—were revoked. The respondent’s flexible working arrangements policy (the “FWA Policy”) accordingly provided that educational support staff were ineligible for flexible work arrangements during term time. These changes had apparently taken place before the claimant’s cancer diagnosis, though the respondent could not identify the precise date.

29 Despite those changes, the respondent’s high school principal, Ms HSP, testified that Mr LM had been “under the impression that flexible working arrangements continued to be in place for administrative staff”, and so the claimant had been permitted to continue working from home on Fridays. I note at this juncture that Mr LM was not called to give evidence in these proceedings. Ms HSP was therefore testifying as to her understanding of Mr LM’s beliefs, derived from what he had apparently told her.

30 As for the additional day, Ms HSP testified that upon the claimant’s return to campus in August 2025, Mr LM “wasn’t really in a position to refuse” her request. Ms HSP explained that “our culture is quite trusting—we want to look after our staff and support them”, and she therefore did not think Mr LM “felt that he could say no”. Ms HSP also testified that the arrangement was entirely ad hoc: that neither day was fixed, and that the claimant would simply notify Mr LM at short notice on whichever days she chose not to come in. This account, however, was again derived from what Mr LM had told Ms HSP rather than from Mr LM’s own testimony. I therefore preferred the claimant’s evidence on this point, though it is in any case not material. The common ground is that the claimant worked from home on two days a week from August 2025, and that Mr LM neither objected to this arrangement nor withheld permission during that period.

31 This arrangement persisted throughout August and September 2025. But beneath the surface, Mr LM was apparently uncomfortable with it. He did not say so to the claimant—at least not directly, and not at first. Instead, he raised his concerns internally, voicing them to Ms HSP and to the respondent’s human resource manager, Mr HRM, both of whom testified about those conversations at the hearing.

32 Mr HRM testified that he had been directed by the respondent’s chief human resource officer to contact Mr LM following concerns that Mr LM had escalated. Mr LM told Mr HRM that the claimant’s work-from-home arrangement was causing disruption to the department, and he wanted to know what options were available to him.

33 Mr HRM’s view was that the claimant was ineligible under the FWA Policy to work from home during term time, given that she was in an educational support role. It was also at this point that Mr HRM turned his attention to the memoranda the claimant had provided from Dr MO and Ms OT. He formed the view that they were insufficient—that they did not clearly define what the light duty recommendation entailed, nor did they specify the duration of the arrangement or the particular accommodations required. Notably, in Mr HRM’s view, the memoranda said nothing about the claimant being required to work from home or to attend twice-weekly rehabilitation sessions. It was this gap, he considered, that meant the respondent had no medical basis for the accommodations that Mr LM had been allowing.

34 Ms HSP’s evidence was to similar effect. She testified that Mr LM had told her, in one of their regular meetings, that the administrative running of the department was becoming difficult because he could not plan for the days when the claimant was not around. The claimant’s role was the first point of contact for any student, parent, or teacher entering the music department, and her physical presence at the front desk was essential to its smooth operation. During her absences, teachers had been pulled away from their core responsibilities to cover administrative tasks, assistance had to be sought from other departments, and day-to-day operations were disrupted. In Ms HSP’s assessment, the arrangement was simply not sustainable from an operational perspective. She therefore decided to convene a meeting with Mr LM and the claimant.

35 In the meantime, Mr LM had begun having his own conversations with the claimant. The first took place on 11 September 2025, at a performance management meeting. According to the claimant, Mr LM said he had no concerns about her administrative performance. But towards the end of the meeting, he asked when she foresaw herself returning to the office full-time—the first time he had expressed any desire for her to be back five days a week. The claimant explained that she was still experiencing the side effects of her treatment and needed more time to recover. She suggested that Mr LM might approach human resources about whether a temporary cover arrangement could be put in place on the days she worked from home, as she understood had been done in other departments.

36 The claimant then asked Mr LM directly what his concern was. His answer, according to her, was that there was no one to replenish the photocopier paper when she was not in. The remark took the claimant aback. It was already a source of distress that she had been unable to carry a ream of paper since her return—the tingling and numbness in her fingers made it impossible to grip one—and she could not understand why her colleagues in the department could not simply replenish the paper themselves, even if it was not strictly their responsibility to do so. She became emotional, and the meeting came to an end.

37 The second conversation was on 2 October 2025. Mr LM told the claimant that he had raised the suggestion for temporary cover with the human resources department but had received no response. He said he intended to arrange a meeting with Ms HSP to discuss working arrangements during the claimant’s recovery. The claimant understood this to mean that the meeting would be about finding ways to cover her work on the days she was not in the office—that Ms HSP would be brought in to understand the situation and perhaps approach human resources to explore what could be done.

38 That meeting with Ms HSP and Mr LM took place on 7 October 2025 (the “7 October Meeting”). The meeting gave rise to four matters of significance.

39 First, rather than any discussion of possible working arrangements—which was the claimant’s understanding of the meeting’s purpose—Ms HSP informed the claimant at the outset that she was required to work from campus Monday to Friday, from 7.45am to 4.45pm, with immediate effect commencing the following day. The basis given was that her educational support role was ineligible for flexible working arrangements under the FWA Policy. The sudden and immediate revocation took the claimant by surprise. She attempted to ask whether she might be permitted to work from home at least one day a week, but there was no room for negotiation.

40 Second, the claimant was informed that the respondent did not consider the medical documentation she had earlier furnished to be adequate. The essential point put across by Ms HSP was that the claimant had not provided any “medical certificate.” That point is best captured in an email Ms HSP sent to the claimant shortly after the meeting, which summarised the discussion:

41 The respondent appeared to be assessing the credibility and adequacy of medical documentation by reference to its provenance and form. Ms OT’s memorandum was, in the respondent’s view, not quite good enough because she was an occupational therapist rather than a medical doctor. And although Dr MO was unquestionably a doctor, and his memorandum had recommended light duty work as Ms HSP acknowledged, it was not a “Light Duties Medical Certificate”.

42 Third, Ms HSP raised the possibility of redeploying the claimant to another role that might be more suitable during her recovery. This suggestion, however, was not explored in any meaningful way. Ms HSP did no more than invite the claimant to approach the human resources department to discuss the possibility if she wished to do so, and the matter was left there.

43 Finally, the claimant was directed at the 7 October Meeting to attend a fit-for-work assessment (“FFW Assessment”) at a private clinic arranged by the respondent. The claimant raised concerns. Dr MO had already assessed her as fit to return to work in the First Oncologist Memorandum that had been provided to the respondent, and she asked why she had to be separately examined by a general practitioner at a private clinic—one who was not familiar with her medical history and who might not be equipped to assess her condition properly. Ms HSP did not respond to those concerns. After the meeting, she sent the claimant a link to the respondent’s fit-for-work policy (the “FFW Policy”), which conferred on the respondent a broad discretion to direct employees “to attend a medical assessment to ascertain that staff member’s ability and suitability to fulfil the inherent role requirements”.

44 The claimant alleged that following the 7 October Meeting, Mr LM began closely monitoring her arrival time, lunch break, and departure time. Mr LM never said as much to her directly, but the claimant drew that inference from his behaviour: she noticed him checking his watch on some occasions at or around the time she reported to her desk, including after lunch. The respondent denied that any such monitoring had taken place. Based on the claimant’s suspicions alone, I was not convinced that any such monitoring had taken place.

45 Despite the claimant’s initial reluctance, the FFW Assessment proved to be in her favour. It took place on 21 October 2025. The clinical notes of the examining doctor, Dr FFW, shed important light on the claimant’s condition at that time. On functional capacity, she was unable to lift loads of five kilograms, and her grip strength in both hands stood at 50% of the norm. On fatigue, Dr FFW recorded that she had “issues of fatigue and requires a period of short rest after eg 2-3 hours of work and occ[asional] naps at lunch break for 10 min”. The notes also recorded that the claimant had been waking at 5.45am to perform the exercises and stretches prescribed as part of her rehabilitation programme, and that this had produced “marked improvement in musculoskeletal functional capacity”. At the same time, Dr FFW noted that her attendance at the office five days per week “has been challenging.”

46 Dr FFW concluded that the claimant was fit to return to work, but he recommended that for an initial two-month period, subject to further review, she should not carry loads exceeding four kilograms and should work a four-day week with one day from home to facilitate her rehabilitation programme and manage her fatigue. He would then review the claimant in three months’ time “for gradual increase in duties towards full [return to work].”

47 These clinical notes and recommendations were sent by Dr FFW to the respondent’s human resource executives on the same day as the FFW Assessment. Rather than accepting the recommendations, however, one of the executives wrote back a few days later, on 24 October, with what was in substance an invitation to reconsider them:

48 Dr FFW responded with a clarification. He reiterated that the arrangement was limited to two months, after which he would review the claimant with a view to stepping up her duties, and noting that her ongoing therapy had proven beneficial:

49 The human resources executive was not satisfied with this clarification and pressed further still:

50 Dr FFW eventually yielded and withdrew his recommendation:

51 Finally, the human resources executive forwarded the exchange to Mr HRM:

52 On 31 October, the claimant was informed that Dr FFW had certified her fit for work and had recommended a restriction on heavy loads. She was then invited by Ms HSP to attend a further meeting on 3 November 2025 (the “3 November Meeting”).

53 The 3 November Meeting was attended by the claimant, Ms HSP, Mr HRM, and a colleague whom the claimant had brought along for support. The substance of the discussion was again summarised in an email from Ms HSP to the claimant, which is worth reproducing in full:

54 The first point restated the respondent’s earlier position at the 7 October Meeting: the claimant could not work from home, her role being ineligible under the FWA Policy. At the 3 November Meeting, the claimant was informed that while Dr FFW had recommended one day of working from home, that recommendation was not mandatory and the matter was left to the respondent’s discretion. In response, the claimant expressed that the flexibility of one day from home would support her recovery—in line with Dr FFW’s recommendation—while still allowing her to discharge her responsibilities properly.

55 The second point concerned the light duty restriction on lifting heavy loads, as recommended by Dr FFW. Ms HSP indicated that the respondent was willing to accommodate this restriction on a “temporary basis” and for a “limited period of time”, but that this accommodation would expire on 11 January 2026, before the second term commenced on 12 January 2026. For the restriction to continue beyond that date, the claimant would be required to submit a medical certificate from Dr MO.

56 Yet there remained no meeting of minds as to what this “medical certificate” actually entailed and how the First Oncologist Memorandum fell short of it. The claimant therefore wrote, in an email dated 6 November replying to Ms HSP’s summary of the 3 November Meeting:

57 The claimant also expressed puzzlement as to why her light duty restriction would expire on 11 January 2026, ahead of her scheduled review with Dr FFW on 22 January 2026. In her view, it would be more sensible to await Dr FFW’s review and assessment before the respondent determined that she was fit to work without any medical restrictions.

58 The third and fourth points were of greatest significance. Recall that the claimant had been attending twice-weekly rehabilitation sessions at the SCS rehabilitation clinic. That programme had been noted by Dr FFW to have produced “marked improvement in musculoskeletal functional capacity,” and his recommendation of one day working from home had been made precisely to facilitate her attendance at “ongoing OT”—therapy he described as having “shown to be very helpful in her strength development over time.”

59 Yet, while the claimant would still be permitted to leave work early twice a week to attend her rehabilitation sessions, the respondent imposed a new restriction: she could only do so from 4.00pm onwards, meaning that she could only attend the 5.00pm sessions. Those sessions, however, were available solely on Mondays, Tuesdays, and Fridays, with each session limited to a maximum of five patients. Demand for them was considerable. The claimant testified that prior to the 3 November Meeting, she would generally try to secure 5.00pm slots, but if she could not, she would book 4.00pm slots—which were far more accessible, being available on every weekday.

60 The claimant sought to negotiate permission to attend 4.00pm slots on occasions when she could not secure the more competitive 5.00pm slots. She proposed that if such permission were granted, she would leave work at 3.30pm—just half an hour earlier than the 4.00pm departure time the respondent was prepared to accommodate—and make her own way to the clinic by private hire transport. Ms HSP and Mr HRM did not accede. They said that if she could not secure a 5.00pm slot and could only book a 4.00pm slot, she would be required to apply for half-day leave. Alternatively, she could seek out a different rehabilitation provider altogether that offered 5.00pm sessions. The claimant did not regard this second option as feasible: the sessions at the SCS rehabilitation clinic were provided free of charge, whereas any alternative provider would require her to bear the costs herself.

61 Much of the present dispute turned on the parties’ opposing characterisations of the series of interventions that unfolded between the 7 October Meeting and the 3 November Meeting. The claimant contended that the respondent breached its duty to take reasonable care of her health and safety through this series of interventions. The respondent, for its part, maintained that it had discharged that duty for reasons to be elaborated below.

62 Before evaluating those competing positions, recall first an observation made at the outset of this judgment (see [5] above). An employer holds the right to direct the labour of its employees as it sees fit—their place of work, their working hours, the scope of their tasks. But it is qualified by countervailing duties, among them the duty to take reasonable care of the employee’s health and safety. The former cannot be exercised in disregard of the latter. That observation bears directly on how we ought to understand the operational needs of employers, and the internal policies that are commonly promulgated in today’s workplace to serve those operational needs.

63 Consider, by way of illustration, a construction company that has project timelines and milestone targets defining its operational needs. To this end, that company directs a group of its workers to be present at the worksite and perform their usual tasks every working day. It also issues a policy specifying the circumstances in which a worker may be excused from the default direction to be present at the worksite and perform her or his usual tasks. Medical grounds would naturally feature among those circumstances. A worker might be permitted to be absent from the worksite altogether for illness or injury if she or he is certified unfit for work by a medical doctor for so long as that medical certificate remains in force. Or suppose the worker has sustained an injury that allows her or him to otherwise be present at work, subject to a restriction of carrying heavy loads so as to facilitate recovery and guard against further harm while at work: the policy might require the production of a medical certificate attesting to that restriction.

64 Such policies serve a useful purpose. They provide the employer with a measure of operational certainty. By specifying in advance the circumstances under which a worker may be excused from her or his usual duties—and the documentation required to establish those circumstances—the employer is able to plan and organise its workforce with reasonable confidence. It knows, within defined parameters, how many workers it can expect to have present on any given day, and can make arrangements accordingly to meet its project timelines, milestone targets, and other operational commitments.

65 Such policies provide structure and consistency for employees as well. They establish clear expectations, afford a degree of transparency as to what is required of them, and ensure that requests for accommodation are handled in an even-handed and predictable manner. In this way, a well-crafted internal policy serves both employer and employee: it preserves the operational integrity of the enterprise while at the same time providing latitude for the accommodation of genuine welfare needs.

66 But internal policies, however well-crafted, ought not to fetter the employer’s discretion to act outside their terms where the circumstances demand it. An employer cannot take refuge in the letter of a policy of its own making to escape the reach of its duties. Return to the illustration. Suppose that one of the workers suffers a shoulder injury. There is a memorandum from a doctor explaining the nature of the injury and advising that the worker be placed on light duty work for the time being. The document does not employ the precise language or form that the internal policy prescribes—it is not styled a “medical certificate”; it does not specify that the worker should be excused from lifting heavy loads; it does not state how long the light duty accommodation should last. Could the employer reasonably respond by pointing to the policy, declaring the documentation deficient, and proceeding to require the worker to carry on with the usual heavy lifting tasks regardless, so that there is no risk of delay to project completion timelines? That, in my view, would be unreasonable. This answer, I suspect, would also strike ordinary persons as intuitively obvious, without any need for a legal textbook.

67 The intuition that such a response is unreasonable is readily explicable. Resorting to the language of the law, where an employer knows—or ought to know—that an employee faces a real risk of harm, it cannot absolve itself of responsibility simply by pointing to what its own policy demanded, or the fact that there are operational targets to meet. These must, in such circumstances, yield to the employer’s duty of care. Policies, guidelines and operational targets are instruments of good management. But there will be circumstances that call for departure from them, and in such circumstances they ought not to become a mechanism by which an employer directs its employees to, borrowing Lord Denning’s words from a different context, “follow the wrong path until [they] fall over the edge of a cliff” (Ostime (Inspector of Taxes) v Australian Mutual Provident Society [1960] AC 459 at 489). But we need not confine ourselves to the strict language of legal duty alone. We may appeal, too, to the basic and ordinary considerations that govern human relationships: the imperative or virtue to treat with respect and compassion those who are suffering some plight or vulnerability, whether from illness, injury, or otherwise—even if doing so comes at some cost or inconvenience to oneself.

68 These further observations bear directly on the present case, for the respondent’s defence rested in large part on one of its internal policies—specifically, its policy concerning medical documentation. To its credit, the respondent accepted that if the claimant’s health genuinely required certain adjustments or accommodations, it would be obliged to provide them notwithstanding any operational inconvenience. Ms HSP testified that although the claimant’s physical absence from campus had caused some degree of operational difficulty in the running of the music department, those concerns—and even the terms of the FWA Policy—would have to yield if the claimant’s health and safety genuinely called for it. She went so far as to accept that had the claimant’s condition required her to work from home, a part-timer could have been engaged to cover the front desk. Mr HRM, too, accepted that if the claimant’s health required any accommodations, including the ability to work from home, the respondent owed her a duty to provide them.

69 That professed willingness to accommodate, however, could not be reconciled with the respondent’s conduct. Following the FFW Assessment and Dr FFW’s advice that she be placed on a restriction against carrying heavy loads and permitted to work from home once per week, it may be recalled that the respondent’s HR executive wrote back to enquire whether the advice to work from home was “just a recommendation” or “mandatory”. Mr HRM explained at the hearing that Dr FFW’s initial advice had been phrased in terms of both a “recommendation” and a “restriction”, which had caused confusion: in his view, a “recommendation” was optional and a “restriction” mandatory, and the HR executive was therefore seeking to clarify which Dr FFW had intended. Had Dr FFW confirmed that the advice was mandatory, Mr HRM said, the respondent would have complied.

70 It was, in my view, not possible to read the correspondence as a genuine attempt at clarification—not even on the most charitable interpretation. The HR executive’s questions were framed so as to steer Dr FFW towards a predetermined conclusion—she sought to “confirm that this [was] just a recommendation and not mandatory”, and pressed him to effectively retract the advice on the basis that the claimant “cannot work from home during school term” given that she was “in a student facing role.” The internal email that followed Dr FFW’s eventual concession was equally telling. The HR executive did not record that Dr FFW had “clarified” his recommendation. She wrote that he had “concurred” with the respondent’s position—in other words, a position had been formed within the human resources department even before the exchange began.

71 But even extending the respondent the benefit of the doubt on this point—that is, assuming arguendo that it had been genuinely prepared to accommodate whatever adjustments the claimant’s health required, operational inconvenience notwithstanding—its position remained that the claimant had simply failed to furnish the requisite medical documentation to establish that any such adjustments were necessary. Evidently, Dr FFW’s medical advice, transmitted directly to the inbox of the respondent’s HR executive, was insufficient as medical documentation from the respondent’s perspective. That raised a question of some importance: what, precisely, did the respondent regard as adequate? It is a question to which I shall return.

72 Suffice it to say for present purposes that the respondent’s position taken in this dispute was this: in the absence of medical documentation satisfying its prescribed requirements, it was not obliged to accommodate the claimant in any respect for health and safety reasons, and the standard of care it owed her was no different from that owed to any other employee. On that footing, none of its decisions or interventions in respect of the claimant could amount to a breach of its duty to take reasonable care of her health and safety.

73 This characterisation of the respondent’s approach could be discerned not only from the series of interventions between the 7 October Meeting and the 3 November Meeting, nor from the testimony of its witnesses alone. It was also committed to writing by Mr HRM himself in an email to the claimant dated 7 November 2025 (the “7 November Email”). The sequence of correspondence that gave rise to the 7 November Email warrants brief explanation. Following the 3 November Meeting, Ms HSP had emailed the claimant with a summary of the discussion. The claimant replied on 6 November to restate certain positions she had expressed at that meeting, which she felt Ms HSP’s summary had not adequately captured. The 7 November Email was Mr HRM’s response. It is apposite to set it out at some length, for it expressed the respondent’s position and approach with a candour that merits careful attention.

74 Mr HRM opened by stating that the respondent had not been provided with the requisite medical documentation to support any of the requested accommodations, and that accordingly “there is no obligation… to make any changes to your role”:

75 The email proceeded to acknowledge the memoranda furnished by the claimant from Dr MO and Ms OT, and the medical advice of Dr FFW—three healthcare professionals who had each, in their own way, recommended some form of accommodation in light of the claimant’s condition. Yet the conclusion Mr HRM drew from this body of medical opinion was that the claimant was “deemed fit for work without restrictions from a medical perspective”:

76 Mr HRM then made clear that the respondent was nonetheless willing to recognise the light duty restriction “despite the lack of supporting medical evidence”, but that such accommodation would cease at the commencement of the second term on 12 January 2026 unless the claimant produced a “light duties’ medical certificate” in the form the respondent required:

77 Two points arise from this passage. As to form, Mr HRM subsequently provided the claimant with a sample medical certificate in a separate email dated 19 November 2025—a document headed “Medical Certificate”, stating “LIGHT DUTY LEAVE” with a specified start and end date, and setting out particular restrictions in terms such as “please excuse from duties involving prolonged standing or walking”. It was only at this point that the respondent’s policy as regards medical documentation finally became clear to the claimant. She now understood precisely how the First Oncologist Memorandum fell short: it was headed “Memo” rather than “Medical Certificate”; it described light duty as “recommended” rather than mandated; and it specified no fixed duration.

78 As to provenance, Mr HRM made plain that “any future medical advice provided… should be from your treating doctor only”—that is, Dr MO alone. The respondent would not, it appeared, accept such documentation from any other medical practitioner, including Dr FFW himself. The email then turned to the questions of therapy sessions and working from home, invoking the FWA Policy in respect of the latter:

79 Mr HRM closed on the following note, which laid bare the respondent’s position: it was not required to provide any additional or heightened accommodation for the claimant’s health and safety in the absence of documentation satisfying its prescribed criteria; that it had nonetheless been accommodating in certain respects out of goodwill—goodwill that was, by this point, evidently wearing thin and subject to review at the commencement of the next term:

80 The respondent’s position, reduced to its essentials, was this: unless an employee produced a document titled “medical certificate”, issued by her primary physician, carrying well-defined and mandatory medical directives, the employer was not obliged to accept that the employee had genuine needs or vulnerabilities capable of heightening the practical demands of its duty of care. Absent documentation of that precise form and character, any accommodation extended in response to the employee’s stated needs would be a matter of goodwill alone—in other words, supererogatory, and revocable at will.

81 The respondent’s policy as regards medical documentation had, in this way, come to inform and distort its entire understanding of its duty to take reasonable care. It ought, of course, to have been the other way around: the duty should have shaped the policy, not the reverse. Instead, the respondent had replaced the open-textured, employee-centred duty that the law implies with something far more emaciated—a duty, in effect, to comply with a medical doctor’s directives, and nothing more.

82 That is not what the implied term requires. The duty to take reasonable care of an employee’s health and safety is, by its nature, open-textured and responsive to circumstances. Its practical demands vary with the situation of the particular employee in question. I readily acknowledge that open-textured duties of this kind resist reduction to a tidy set of rules, and hence a degree of indeterminacy is inherent in this duty. But an employer should seldom find itself on the wrong side of the line if it exercises practical judgment by reference to what an ordinarily prudent employer would do, informed by whatever credible information and medical advice is available to it concerning the employee in question. No rational and reasonable employer requires a doctor to draft an operational manual of mandatory directives before it can form a view as to what the health and safety of its own employee reasonably demands. That is a judgment the employer is well placed—and certainly well obliged—to make for itself.

83 In the present case, it was not open to an ordinarily prudent employer to arrive at the conclusion Mr HRM did—that the claimant was “fit for work without restrictions from a medical perspective”. That conclusion was contradicted by the documented medical advice of three separate healthcare professionals, each of whom had identified that the claimant was suffering side effects from cancer treatment, and each of whom had recommended accommodations such as light duty restrictions, albeit with varying degrees of specificity. It was beside the point whether that advice was expressed in the form of a medical certificate, or whether it was couched in mandatory rather than recommendatory language. Nor did the fact that Ms OT was an occupational therapist rather than a medical doctor render her professional opinion as to the claimant’s functional capacity and recovery any less worthy of serious consideration. An ordinarily prudent employer does not parse the form and provenance of medical opinion to avoid engaging with its substance.

84 To arrive at Mr HRM’s conclusion was also to deny the truth of what the claimant had herself shared with the respondent about her medical condition and her needs. To be sure, the existence of opportunistic employees in the labour market requires employers to be circumspect. Yet, there was no suggestion that the claimant was an opportunistic employee seeking to exploit her employer’s goodwill—and indeed, such a suggestion would have been difficult to sustain. While it may not be especially difficult to feign the symptoms of, say, a stomach-ache, feigning the side effects of a cancer diagnosis and its attendant treatment would be quite the feat.

85 To the contrary, the evidence before me painted a picture of a capable and conscientious employee. She had been in the respondent’s employ since 2008, and it was undisputed that Mr LM had recommended her as a “prime candidate” for promotion shortly before her cancer diagnosis, even if nothing ultimately came of it. Where an employer is told by any employee of her or his medical needs and vulnerabilities, it is incumbent upon that employer to reasonably investigate, consider, and respond to the matter with care and sensitivity—which, as I observed earlier, is the very starting point of the duty to take reasonable care in cases of this kind. How much more so, then, for a longstanding employee, with the trust and goodwill of 17 years of service behind her? Mr HRM’s declaration that “there [was] no obligation… to make any changes to your role given there is no medical evidence to support your claims” was, by any measure, a far cry from that standard.

86 It appeared to me, in fact, that the respondent’s insistence on a particular form of medical documentation was nothing more than a pretext—a means of avoiding what it ought to have done, which was to reasonably investigate and consider the substance of the claimant’s health needs with care and sensitivity. The cause of this inertia was not altogether clear to me. Was it the administrative convenience of adhering to the terms of the FWA Policy and making no exception? Or was it that the respondent had allowed its operational priorities to crowd out its duty of care?

87 The answer, in the end, mattered little. For whatever the cause, neither excuse withstood scrutiny in light of the observations made earlier in this judgment: that internal policies and operational priorities must yield, where the circumstances demand it, to the duty to take reasonable care of an employee’s health and safety. This is not to trivialise the respondent’s operational concerns. I accepted that Mr LM genuinely found the department difficult to run during the claimant’s absences; that her days at home were not always predictable; and that her role placed her at the front desk, the first point of contact for anyone entering the music department. The duty does not demand that an employer capitulate to every request at the expense of its legitimate business requirements.

88 What it does demand, as I said earlier, is the exercise of practical judgment. An ordinarily prudent employer, reasonably investigating and considering the claimant’s health issues with care, would naturally take into account the operational requirements of the business. But it would equally have regard to the claimant’s medical status and limitations, the likely trajectory of her recovery, the types of accommodation that her condition might require, the risk to her health if accommodation were refused, the impact of any such arrangements on her colleagues, and whether suitable alternatives existed.

89 None of this can be done without genuine engagement with the employee. An employer that takes no sensible steps to understand a matter of health and safety drawn to its attention can hardly expect to discharge its duty to reasonably consider and respond to it. I should make clear, moreover, that I do not suggest the engagement had to produce any particular outcome in the present case. That was precisely the point of such a discussion—a range of reasonable outcomes might have emerged from it. Perhaps it would have meant one fixed work-from-home day rather than two. Perhaps part-time cover at the front desk, of the kind Ms HSP herself accepted was possible. Perhaps a trial arrangement, subject to periodic review as the claimant’s recovery progressed. Or perhaps consultation would have revealed that no work-from-home arrangement could be sustained during term time, and the parties would have turned to other measures altogether.

90 The respondent made no such attempt. After Mr LM raised his concerns, the respondent formed its views in haste and arrived at the 7 October Meeting already entrenched in them: that the claimant was to be on campus five days a week without exception. There was no room for negotiation, and no genuine willingness to consider or discuss the claimant’s concerns. The respondent dismissed the adequacy of the medical documentation she had already furnished, adopting a stance that was at once sceptical of her needs and indifferent to their substance. And while Ms HSP did raise the possibility of redeployment, it was not meaningfully explored—it was presented, rather, as a problem for the claimant alone to pursue with the human resources department if she wished to do so. In my judgment, this was an unreasonable and insensitive response to the matters of health the claimant had raised, and the respondent was, by this point, already in breach of its duty of care.

91 What followed only deepened the breach. The respondent wilfully rejected Dr FFW’s recommendations and doubled down on its already untenable conclusion that the claimant had failed to furnish adequate medical documentation of her needs. It then issued her a deadline to produce the prescribed documentation by the commencement of the next term, or be treated as any other reasonably healthy employee. That erroneous conclusion emboldened the respondent to tighten its grip further, placing additional obstacles to the claimant’s attendance at her rehabilitation sessions, even as the medical documentation before it made plain how vital those sessions were to her recovery. The irony was that, all the while, the respondent characterised this conduct as manifestations of its goodwill. It was, in truth, anything but. It was a course of conduct completely lacking in the sensitivity and care that common decency requires of any person dealing with another in vulnerability—let alone the standard that the law demands of an employer charged with the duty to take reasonable care of its employee’s health and safety.

92 Taken together, the respondent’s conduct sent a consistent and unmistakable message to a long-serving employee of 17 years who had survived stage 3 cancer and returned to work: that her medical needs and vulnerabilities were not to be taken seriously, and that the respondent was not willing to discharge its duty to take reasonable care of her health and safety. Yet the benefits of employment—the wages, the financial security, the professional life that work affords—can only be meaningfully enjoyed by an employee who is able to attend and perform her duties without her health being put at risk in the process. An employer that persistently refuses to take reasonable care of a medically vulnerable employee strikes at something essential to the employment relationship itself. I was therefore satisfied that the respondent had not only breached its duty to take reasonable care of the claimant’s health and safety, but that the breach was a fundamental one whose nature and consequences were so serious as to “go to the root of the contract”: see RDC Concrete Pte Ltd v Sato Kogyo (S) Pte Ltd [2007] 4 SLR(R) 413 at [99].

93 For similar reasons, I was satisfied that the respondent had also breached the implied term of mutual trust and confidence. That term imposes an obligation that neither party to an employment contract shall, “without reasonable and proper cause, conduct itself in a manner calculated and likely to destroy or seriously damage the relationship of confidence and trust between employer and employee”: see Malik v Bank of Credit and Commerce International SA (in compulsory liquidation) [1997] IRLR 462 at [54].

94 In my view, an employer’s fundamental breach of its duty to take reasonable care of its employee’s health and safety would have the effect of undermining the trust and confidence upon which the employment relationship depends—for health is the very foundation upon which an employee’s ability to work, and to derive the benefits of work, rests. An employer that undermines that foundation can hardly be said to be conducting itself in a manner consistent with the trust the employment relationship requires. Indeed, the High Court in Cheah Peng Hock v Luzhou Bio-Chem Technology Limited [2013] 2 SLR 577 took the view that the employer’s breach of the duty to take reasonable care of its employee’s health and safety in Austin constituted a breach of the implied term of mutual trust and confidence (at [56(h)]).

95 My findings on the breach are borne out by what followed. The claimant testified that the respondent’s successive interventions had visited upon her considerable stress, exacting a serious toll on her health—physical, mental, and emotional. She had difficulty sleeping and wept every day. She developed severe headaches and debilitating anxiety, and her jaw began to lock. She was also losing weight. Before her cancer treatment, she had weighed 61.5 kilograms; by the end of treatment, she was down to 54 kilograms. By early November 2025, despite being on the road to recovery, her weight had fallen further still, to 51.5 kilograms.

96 Her first review consultation with Dr MO since returning to campus took place on 7 November 2025. She explained to him the respondent’s demands for further medical documentation and described her deteriorating condition. Dr MO issued her a medical certificate covering the period from 7 to 14 November 2025.

97 He also issued a further medical memorandum that same day (the “Second Oncologist Memorandum”), which elaborated on her condition in greater detail:

98 It should be immediately apparent that the Second Oncologist Memorandum was not in the form the respondent had been insisting upon. At that point, the claimant had not yet been furnished with the sample medical certificate, but she had concluded that providing it would be a futile exercise — and that assessment proved accurate. When asked at the hearing whether the memorandum would have made any difference after the 3 November Meeting, Mr HRM answered that the oncologist had once again recommended light duty “without a definition or a duration”. It would not have moved the respondent at all.

99 Again, I could not accept that conclusion. The Second Oncologist Memorandum stood as even clearer evidence of the claimant’s actual condition—the severe peripheral neuropathy in her hands, and the ongoing rehabilitation she required to cope with it. The vulnerabilities she had been describing to her employer throughout were plainly and repeatedly documented by healthcare professionals. That Mr HRM and his colleagues would remain so fixated on the form of the medical documentation as to miss the unmistakable picture that the substance of that documentation had painted of the claimant’s condition and her needs was difficult to comprehend.

100 The respondent also drew my attention to a line in the medical certificate issued by Dr MO on 7 November, which read “Fit for light duty from N.A. to N.A.”, and relied upon it to reinforce its position that even Dr MO himself had not certified the claimant’s need for light duty restrictions. This submission merits only brief treatment. This medical certificate was issued primarily to certify that the claimant was unfit for duty from 7 to 14 November, in support of her application for medical leave during that period. For the light duty advice, Dr MO had by then already issued two memoranda—albeit that only one had been furnished to the respondent. To treat a clerical “N.A.” notation as negating the considered professional opinions of Dr MO, Ms OT and Dr FFW—all of whom had recommended light duty accommodations for a person suffering the long-term side effects of cancer treatment—would be a serious error of practical judgment. It therefore did not assist the respondent’s case in the slightest.

101 In the weeks that followed, the claimant took extended periods of medical leave, citing her deteriorating health as the reason. After her initial leave ended on 14 November, she took further medical leave on 20 November, from 25 to 27 November, from 1 to 15 December, and from 16 December 2025 to 16 January 2026. A telephone call with her oncologist on 8 December 2025 proved to be a turning point. According to the claimant, he told her that her reported symptoms were “very concerning” and that she needed to take time off work. It was then, she said, that she came to understand the cumulative toll that the work environment had taken on her health, and that she needed to resign in order to protect it.

102 She therefore tendered her resignation on 19 December 2025. In her resignation email, she wrote:

103 She received an acknowledgement from a HR executive and was informed that the offboarding process would be initiated. Her one month’s notice expired on 18 January 2026, which was the last day of her employment.

104 On that final day, she wrote to the respondent’s leadership team, expressing her “concern and disappointment” regarding the handling of her situation by the school leadership during “a very vulnerable period of [her] life, following [her] return to work after severe cancer treatment”. She recounted her long service and contributions to the respondent since 2008—memories which no doubt were fond ones—but said that her “recent experiences ha[d] left [her] feeling unsupported and unfairly treated”, which was “both distressing and disheartening” for her. She elaborated that her “request for temporary accommodations—light duties, flexible working arrangements, and time for medically necessary therapy sessions—[had been] met with resistance and unclear communication”. She questioned “the motivations behind the handling of [her] employment and the school’s commitment to supporting staff through medical difficulties”.

105 The email was, on the whole, politely written. The claimant made clear that her purpose was “for the college to be fully aware of how local staff [were] managed, especially during times of vulnerability”, and so that “future staff in similar situations receives better support and understanding”. She received no reply or follow-up from the leadership team.

106 On 11 February 2026, she submitted a mediation request under s 3(1) of the Employment Claims Act 2016 (2020 Rev Ed) on the grounds that she had been dismissed without just cause or excuse under s 14(2) of the EA. The matter was referred to a mediator under the Tripartite Alliance for Dispute Resolution. It never proceeded to mediation, however. A claim referral certificate was directly issued by the mediator on 18 March 2026 on the ground that she was satisfied—for unstated reasons—that there was no reasonable prospect of settling the dispute through mediation. The respondent had only been notified of the claimant’s raising of the dispute on the same day that the claim referral certificate was issued. The claim was then lodged in the Employment Claims Tribunals on 30 March 2026.

107 It may, on its face, seem incongruous that the claimant’s claim was that she had been dismissed without just cause or excuse under s 14(2) of the EA, when it was undisputed that she had resigned. But the incongruity dissolves once one appreciates that a resignation may, in certain circumstances, be treated at law as a dismissal—what is commonly called “constructive dismissal”, a dismissal by legal construct.

108 Both parties referenced this concept in their submissions. The claimant alleged that it applied; the respondent denied it. But the term admits of more than one conception, and it is important to be precise about which conception governs before turning to the facts.

109 There is the familiar common law conception. The Court of Appeal in Wee Kim San Lawrence Bernard v Robinson & Co (Singapore) Pte Ltd [2014] 4 SLR 357 (“Wee Kim San”) at [23] explained that the term:

110 Put simply, the common law conception is rooted in the law of contract: the employer commits a repudiatory breach which entitles the employee to treat the employment contract as discharged, and the employee accepts that breach by resigning, with or without notice. In such cases, the law treats the resignation as a termination by the employer.

111 I would, however, be cautious about appealing directly to the common law conception in the s 14(2) context. The reason is that s 2(1) of the EA sets out its own statutory definition of “dismiss”, and that definition contains within it a conception of constructive dismissal that is expressed in different terms:

112 The statutory conception, as can be seen, does not speak the language of repudiatory breach and the employee’s entitlement to treat the contract as discharged. Instead, it turns on the voluntariness of the resignation—specifically, whether the employee was forced to resign by the employer’s conduct. This is a departure from the common law conception (see also Prashant at [146]–[147]).

113 The difficulty, however, is that the language of “voluntarily” and “forced” in the statutory definition is not altogether straightforward. Concepts of voluntariness and compulsion bear different meanings across different legal contexts, shaped by the purposes those contexts serve. I shall not attempt to lay down a fully developed account of how these concepts should be understood in the context of the statutory definition—that is a task better undertaken as cases accumulate and the contours of the provision are tested. But it remains necessary for me to make some brief observations on the meaning I believe these concepts should bear, being the conceptions applied in the present case.

114 First, to say that an employee resigned involuntarily because she was “forced” to do so cannot mean that she was deprived of the physical and psychological capacity to do otherwise. That conception of involuntariness may have its place in the criminal law, where the voluntariness of an act goes to whether the accused can be said to have acted at all for purposes of the actus reus. But an employee who resigns always (save in perhaps the most exceptional of cases) retains this capacity. If “forced” meant the complete negation of that capacity, then no resignation could ever qualify, however intolerable the circumstances that prompted it. Surely this was not Parliament’s intention.

115 Nor does “forced” require that resignation was absolutely necessary in the sense that no other course of action was conceivable. However oppressive an employer’s conduct, the employee will always have options—to endure, to seek redress through other means, to negotiate. If the statutory test demanded that all such alternatives be foreclosed before a resignation could be treated as involuntary, it would again be rendered largely ineffective. That, too, could not have been the legislative intent.

116 The better view, it seems to me, is that an employee is “forced” to resign—and therefore does not do so “voluntarily”—where the employer’s conduct has left the employee with no practical alternative or choice but to resign, assessed by reference to what it was reasonable to conclude in the circumstances. This is an objective inquiry, but one that must be conducted with regard to the employee’s particular situation—in the present case, the claimant’s medical condition, her vulnerabilities, and the pressures she faced are not irrelevant background noise but part of the very context against which the reasonableness of her conclusion must be measured. It is not enough, of course, that the employee subjectively felt that she or he had no choice, however deeply felt that conviction may have been. The question is whether the choice was constrained such that a person in the employee’s position could reasonably have concluded that there was no practical alternative but to resign.

117 Then there is, finally, the question of causation. The statutory definition requires not merely that the employee lacked a practically reasonable alternative, but that this state of affairs was brought about by the conduct or omissions of the employer—in the sense that the employer’s conduct was the operative cause of the resignation.

118 With these principles in mind, I was satisfied that the claimant’s resignation was not voluntary and that she was forced to resign as a result of the respondent’s course of conduct beginning from the 7 October Meeting.

119 I had found that this course of conduct amounted to, first, a fundamental breach of the respondent’s duty to take reasonable care of the claimant’s health and safety; and second, a breach of the implied term of mutual trust and confidence—which is necessarily a repudiatory breach of contract: see Morrow v Safeway Stores plc [2002] IRLR 9 at [23].

120 Although I observed earlier that the operative conception of constructive dismissal under the EA is not framed in the language of the employee’s entitlement to treat the contract as discharged due to the employer’s repudiatory breach, that does not render findings of repudiatory breach altogether irrelevant to the statutory inquiry. Where the nature and consequences of a breach are sufficiently serious to permit the employee to terminate the contract under the general law—as I had found them to be here—I struggle to imagine a situation in which it would still not be reasonable for the employee to have regarded herself as having no practical alternative but to resign. The two analyses seem to converge on the same underlying reality, being that the respondent’s conduct had rendered the continuation of the employment relationship, in the manner the contract implicitly envisages, no longer reasonably possible.

121 I am mindful of the possible objection that an employee faced with a power to discharge the contract for the employer’s repudiatory breach retains the option to affirm the contract—and that the theoretical availability of this choice might be thought to preclude any finding that the employee was “forced” to resign. I do not find this line of reasoning persuasive. As I explained earlier, the mere theoretical availability of an alternative is not, by itself, sufficient to negate a finding that a choice was made under compulsion. The law recognises this in other contexts. For instance, doctrines of duress in contract and criminal law recognise that a person may be relieved of the legal consequences of a decision reached under pressure, even though that person could, in theory, have resisted and chosen to suffer the threatened consequences rather than submit. Yet, the law looks past the formal availability of a choice to the practical reality of the constraints under which it was made. An employee who resigns in the face of an employer’s repudiatory breach (rather than affirm the contract) stands in an analogous position—she or he can reasonably conclude that there was no practical alternative but terminate the contract because of the employer’s repudiatory breach. But I am conscious that my imagination may not extend to every case in which the two analyses might diverge, and I leave the broader question open concerning the relationship between the common law conception of constructive dismissal and that which operates under the EA.

122 On the facts of this case, the respondent’s persistent failure of duty placed the claimant in an unenviable dilemma: she could remain in employment and continue to expose herself to the risk of further deterioration to her health, or she could resign and protect it. That was a constrained choice thrust upon the claimant by the breach of the respondent’s obligations, and the claimant quite reasonably chose her health over her employment. She did not therefore resign voluntarily but was forced to do so, and that her resignation accordingly amounted to a dismissal within the meaning of the EA.

123 The fact that a resignation may be treated at law as a dismissal does not, however, automatically mean that the dismissal was without just cause or excuse under s 14(2) of the EA. Put another way, it is entirely possible for an employee to have resigned involuntarily—and hence to be properly regarded as dismissed—and yet for the employer to have had just cause or excuse for that dismissal. The two questions are distinct, and they must be addressed separately.

124 An illustration makes the point. An underperforming employee may have undergone, and failed, a robust performance improvement process, whereupon the employer informs the employee that it intends to dismiss him but offers the option to resign instead—a reasonable proposal to spare the employee the stigma of a formal dismissal on his employment record. The employee’s resignation in such circumstances may well constitute a dismissal within the statutory definition, yet the employer may nonetheless be able to establish just cause or excuse by reference to the very capability concerns that prompted the process in the first place.

125 The upshot is this: even where an employee has been constructively dismissed within the EA’s conception, a claim under s 14(2) requires a separate examination of whether the employer had just cause or excuse to dismiss the employee at the material time — that is, at the time of the resignation. I therefore turn to that question on the facts of the present case. Would there have been just cause or excuse for the respondent to dismiss the claimant on the ground of her medical condition and needs?

126 The starting point is that the claimant had been assessed fit to return to her job by both Dr MO and Dr FFW. Both assessments were, in my view, credible. Dr FFW had been provided a copy of the claimant’s job description; and it was reasonable to assume that Dr MO, as the claimant’s treating oncologist, had taken a detailed social history of his patient that would have encompassed her employment particulars. Their medical opinions indicated that she was fit to perform the administrative and secretarial functions of her role, and that discharging those functions would not—subject to certain guardrails—put her health and safety at risk.

127 Those guardrails existed because, while the claimant was fit to return to work, she was by no means in a state of complete recovery. An ordinarily prudent employer making reasonable adjustments and accommodations, having regard to her condition, her needs, and her vulnerabilities, would undoubtedly do so at some cost and disruption to its operations. She had to take frequent breaks. She had to attend twice-weekly rehabilitation sessions. Proper consultation with the claimant may also have revealed that some form of flexible working arrangement was necessary. The claimant herself readily acknowledged that she was, on the whole, a less productive employee.

128 This reality carried practical consequences. Operational tasks ordinarily handled by the claimant would be discharged at a slower pace. She might not be at the front desk at all hours or on all days that her employer would otherwise have preferred. Her absences and reduced capabilities might, in turn, become an obstacle to meeting operational standards—having someone at the front desk at all times, or someone to replenish photocopier paper when it ran out—or might require cumbersome or costly operational adjustments, such as engaging a part-timer on days when the claimant was not present. Her colleagues might find themselves shouldering additional responsibilities; teachers in the department, for instance, might have to handle administrative tasks for which they would otherwise have had the claimant’s support. And there was no fixed timeline for the claimant’s recovery from the side effects of her cancer treatment, so the likely duration of these accommodations could not be foreseen with any measure of certainty.

129 In such circumstances, it might be tempting for an employer in the respondent’s position to dismiss the claimant as a quick-and-easy solution and hire someone else in her stead—although I should clarify that there was no evidence to suggest that the respondent had been considering this option. An employer is of course contractually entitled to terminate the employment contract by giving due notice or salary in lieu thereof, but such contractual compliance does not quite address the statutory inquiry of whether the dismissal was with or without just cause or excuse under s 14(2) of the EA.

130 In this regard, while dismissal on the ground of an employee’s medical condition and needs may certainly constitute just cause or excuse under s 14(2), whether that threshold is reached depends very much on the facts. This is, again, not a question that yields to a tidy set of rules producing determinate answers. What matters are the approach an employer takes, and the considerations it brings to bear, when navigating the question of how to respond to an employee with a medical condition and needs. I offer some brief observations on each.

131 As to approach, an employer should navigate this issue with care, sensitivity, and respect for its employee as a human being. It may seem superfluous to underscore the humanity of the employee, but the calculus of performance targets, operational convenience, or institutional self-interest may sometimes displace this fact from the minds of those who make decisions within an employing organisation. That humanity means that an employee with a medical condition or needs requiring accommodation should not be viewed as a broken machine to be written off from the organisation’s human inventory and replaced by a healthier procurement. To dismiss such an employee as a first resort is to make her or him feel precisely that way—that the employee is somehow diminished, or of lesser worth by reason of her or his condition.

132 Moreover, decision-makers must of course appreciate that whether and when any of our lives are touched by illness or affliction, and of what kind or severity, is something none of us can know in advance. What we do know, from ordinary life and experience, is the readily observable toll that serious medical conditions exact on those who suffer them. It is not only the physiological effects; sometimes it is the mental and emotional anguish that proves the more intolerable burden. That anguish may come from many sources: the struggle to make sense of one’s suffering; the perceived helplessness about one’s circumstances; the uncertainty of recovery; the dejection that comes from comparing oneself with healthier times; the despondency of a prognosis that offers little hope; the fear of being a burden on others and of losing one’s livelihood.

133 An employee with a medical condition and needs may face such challenges, and hence the importance—for those around her, including colleagues and managers in the employing organisation—of not adding to that anguish by treating the employee as a problem, a burden, or an imposition to be swiftly resolved. The right course is to treat her or him with appropriate care, sensitivity, and respect, by genuinely considering whether it is realistic and practicable for the employee to continue performing her or his duties satisfactorily despite her or his medical condition and needs.

134 The starting point of any such approach, then, is a reasonable inquiry into the relevant considerations, undertaken before any decision is made. Those considerations include the actual state of the employee’s medical condition and its likely duration, the effect of that condition on the employee’s ability to perform her role, the employer’s operational requirements, the availability of alternative roles or arrangements suited to the employee’s condition, and whether adjustments or accommodations can reasonably be made.

135 For that inquiry to be meaningful, the employee should be consulted for her views. The observations of the EAT in East Lindsey District Council v G E Daubney [1977] IRLR 182 (at [18]) are instructive in this regard. Although those observations were made in the context of the United Kingdom’s unfair dismissal regime, they apply with equal force here:

136 It is only by inquiring into these matters adequately—and consulting the employee as part of that process—that an ordinarily prudent employer can properly decide, on the evidence, whether the employee can satisfactorily perform her duties within a reasonable time; and if not, whether there are alternative arrangements available, or reasonable adjustments or accommodations that can be made, or whether dismissal is warranted in the circumstances. The quality of that inquiry is therefore central to the assessment of whether any resulting dismissal was with just cause or excuse.

137 Measured against that standard, the respondent’s conduct falls short. As I found earlier, it failed to genuinely consult the claimant’s views, and it declined to engage with the available medical advice on the basis that none of the documentation provided constituted a medical certificate. In my view, an ordinarily prudent employer would have periodically reviewed and consulted the claimant on her medical condition, her needs, and the progress of her recovery, while providing reasonable adjustments and accommodations as required. The available medical evidence at the material time pointed toward an employee whose condition appeared to be on a trajectory of improvement. That employer would have observed—consistent with Dr FFW’s assessments and the claimant’s own account—that the rehabilitation sessions at the SCS rehabilitation clinic, combined with the claimant’s discipline in performing her prescribed exercises and stretches each morning, had produced measurable progress. As those improvements accumulated, some of the adjustments and accommodations made could correspondingly be dialled back.

138 The claimant’s own evidence at the hearing confirmed that this trajectory was a real one. She testified that although her health had deteriorated following the series of interventions by the respondent that ultimately informed her resignation, she had since made considerable progress in her recovery. By the time of the hearing, she no longer experienced the same levels of fatigue and no longer required a short nap during her lunch breaks. The neuropathy in her hands—which had at one point prevented her from carrying even a ream of photocopier paper, or peeling the skin off a piece of fried chicken—had improved significantly through the rehabilitation programme. She was still attending twice-weekly rehabilitation sessions, but she expected the frequency to be reduced to a monthly basis after her next review, as the SCS rehabilitation clinic would aim to graduate patients upon their demonstrating a sufficient level of recovery. When asked whether, in her current state, she would still require a flexible working arrangement, she said: “I think I could work from the office every day now.”

139 Granted, that improvement was never guaranteed. But it is that uncertainty which counsels against haste. The absence of a guaranteed outcome was a reason for patience and continued engagement. An ordinarily prudent employer would have allowed reasonable time to pass, periodically reviewing and consulting the claimant on her condition and the trajectory of her recovery, all while discharging its duty to take reasonable care of her health and safety.

140 For the reasons given, I was satisfied that the claimant had been dismissed without just cause or excuse under s 14(2) of the EA. On the question of remedies, I allowed the claim for $20,000 in full, representing almost four months of her monthly salary of $5,080. That award comprised the following components, having regard to the Second Schedule to the Employment Claims Regulations 2017: $15,240, being three months’ salary, for loss of income; and $4,760 for harm.

141 On the loss of income component, the claimant had remained unemployed from her last day of employment through to the date of the hearing, having chosen to focus on her recovery. That was, in my view, an entirely reasonable course of action given her medical condition and needs. In assessing her loss, I took into account the fact that but for the respondent’s course of conduct—which had forced her resignation—she would in all likelihood have remained in employment. The loss of income she suffered was therefore a direct and foreseeable consequence of the respondent’s conduct, and three months’ salary was an appropriate measure of that loss.

142 On the harm component, I would have been minded to award a base amount of two months’ salary, with an upward adjustment of 50%—that is, one additional month’s salary, yielding three months’ salary in total—on account of my finding that the respondent’s course of conduct amounted to a fundamental breach of its duty to take reasonable care of the claimant’s health and safety, and had, in the circumstances, resulted in a deterioration of her health in November and December 2025. That would have produced a harm award of $15,240. The monetary limit of the tribunal’s jurisdiction, however, constrained the total award to $20,000, with the result that the harm component was necessarily limited to the residual sum of $4,760 after the loss of income component had been accounted for.

143 Finally, on the question of costs, I awarded the claimant $400 in costs and $70 in disbursements.

144 I end with a closing observation. At the hearing, Ms HSP told this tribunal that the respondent’s culture was one in which “we want to look after our staff and support them”. Those words were a profession of the belief of the respondent as an organisation. But beliefs, it has been said, are evidenced in our actions. When one turns from Ms HSP’s words to the respondent’s conduct, a considerable distance between the two is revealed.

145 An employee of 17 years, still recovering from a serious illness, was failed in the respondent’s obligation to take reasonable care of her health and safety. That failure eventually drove the claimant to resign. What was lost was not merely a career, long and loyal as it was, but also the many relationships forged in the ordinary course of working life. That all of this should have ended as it did—certainly not in the warmth of the culture Ms HSP described—is regrettable. It is a reminder that the obligation to look after one’s employees with reasonable care is not merely an aspirational belief to be expressed in a written policy or in a courtroom submission. It must find expression in the small but consequential decisions that accumulate over the course of an employment relationship, even when the cost of doing so makes it tempting to treat that aspiration as nothing more than words.