[2022] SGHC 88

The Appellant has a diploma in Nursing and has been enrolled as a nurse with the Singapore Nursing Board since 2002. She was a freelance nurse at the material time and advertised her babysitting services on the Internet sometime in 2016, stating therein that she was a nurse. She resided in an apartment in Hougang with her two teenage daughters.

Sometime in October 2016, the mother of BB1 (“M1”) was looking for babysitting services for her children, and was contacted by the Appellant. M1 started sending over BB1 and her elder daughter of five years old to the Appellant’s home for babysitting sometime in early November 2016. Generally, she would send them over at around 7.30am and pick them up at around 6pm later that day.

The Appellant babysat BB1 and her sister on eight occasions between 7 November and 9 December 2016. The Appellant was the primary caregiver to BB1 during these times and would feed her, including personally preparing her milk. This was save for one occasion where one of the Appellant’s daughters did so. It was agreed that the Appellant’s daughters did not administer any poisons or any stupefying, intoxicating or unwholesome drugs to BB1 during these babysitting sessions.

On 22 November 2016, M1 brought BB1 to KK Women’s and Children’s Hospital (“KKH”) for a check-up, as she noticed that BB1 was behaving unusually. KKH took a blood sample from BB1 and subsequently discharged her as nothing unusual was found. The blood test did not check for any medication or drugs in BB1’s blood.

On 9 December 2016, after picking BB1 up from the Appellant’s home, M1 noticed that BB1 was unnaturally drowsy. She brought BB1 to Parkway East Hospital (“PEH”), whereupon BB1 was admitted. A medical memo dated 16 February 2017 stated that she was found to be drowsy, floppy and unable to follow objects. A blood sample taken from BB1 on 9 December 2016 was sent to the Health Sciences Authority (“HSA”) for analysis on 12 December 2016, and was found to contain the following substances:

BB1 responded well to treatment and was subsequently discharged on 13 December 2016. She was found to be well at a review on 17 December 2016.

As for BB2, sometime in December 2016, her mother (“M2”) posted in a Facebook group called “Mummy’s Group” that she needed someone to look after BB2 on the night of 25 December 2016. The Appellant responded to this message, and it was agreed that M2 would send BB2 to the Appellant’s home for babysitting. M2 subsequently sent BB2 to the Appellant’s home on the said night. Other than the Appellant and her daughters, Dr Peter Looi (“Dr Looi”) was also present in her home. However, he was not involved in BB2’s care. As with the charge involving BB1, it was agreed that the Appellant’s daughters did not administer any poisons or any stupefying, intoxicating or unwholesome drugs to BB2.

On 26 December 2016, sometime between 6.30am and 7.30am, BB2’s father (“F2”) picked BB2 up from the Appellant’s home. M2 sent BB2 to KKH at around 5pm on the same day. BB2 was observed to be drowsy with ptosis (droopy eyelids), truncal ataxia (inability to sit upright and/or stand without support) and had difficulty walking, and was admitted.

A urine sample taken from BB2 on 28 December 2016 was sent to the HSA for analysis the following day. The following substances were detected in the urine sample: hydroxyalprazolam, alprazolam, chlorpheniramine, diazepam, ephedrine, nordiazepam, oxazepam and temazepam. A blood sample taken from BB2 on 29 December 2016 was also sent to the HSA for analysis on the same day, and was found to contain the following substances:

BB2 responded well to physiotherapy and occupational therapy, and was discharged on 1 January 2017.

On 29 December 2016, the police seized the following items from the Appellant’s home:

It was also agreed that in November and December 2016, the Appellant was prescribed, among other things, “Stilnox” (zolpidem), “Xanax” (alprazolam) and diazepam by Everhealth Medical Centre. Furthermore, the following items are available over the counter at pharmacies:

Nordiazepam, oxazepam and temazepam are metabolites of diazepam, while hydroxyalprazolam is a metabolite of alprazolam.

The Appellant was charged with administering alprazolam, chlorpheniramine, (dextro)methorphan, diazepam, orphenadrine, oxazepam, (pseudo)ephedrine, temazepam, triprolidine and zolpidem to BB1 sometime from 7 November to 9 December 2016 at or near the Appellant’s home with intent to cause hurt to her, an offence under s 328 of the Penal Code. She was further charged under s 328 of the Penal Code with administering alprazolam, chlorpheniramine, diazepam, ephedrine, oxazepam and temazepam to BB2 sometime from 25 to 26 December 2016 at or near the Appellant’s home with intent to cause hurt to her.

The Prosecution submitted that the Charges were made out as the Appellant had access to the drugs that were found in the bodies of the Victims: either through prescriptions at the material time, brought over to her home by Dr Looi, or by purchasing them over the counter. It was argued that she administered the drugs to the Victims. The Victims’ mothers had consistently testified that their babies were normal when they were brought over to the Appellant’s home, but unusually drowsy when picked up, and their testimony was corroborated by their messages to the Appellant as well as the fact that they brought the Victims to the hospital to be examined. The poisons were also found in the Victims’ blood samples shortly after they had left the care of the Appellant. The Appellant did not moreover have a satisfactory explanation for the milk bottle with traces of zolpidem which was found in her home. It was argued that the Appellant, being an enrolled nurse and being fully aware of the dangers of administering non-prescribed medication to babies, clearly intended to cause hurt to the Victims by so administering the poisons to them.

With regard to sentence, the Prosecution highlighted the vulnerability of the Victims, and the testimony of Dr Juliet Tan (“Dr Tan”) and Dr Koh Ai Ling (“Dr Koh”), Registrars at the Department of Paediatrics in KKH who had prepared medical reports for BB2, that it would be unsafe to administer the drugs that were found in the Victims without close monitoring. The Prosecution sought a sentence of at least seven years’ imprisonment, bearing in mind that the sentencing norm for such cases was in the range of three to four years’ imprisonment and on an application of the totality principle. Furthermore, where victims of similar vulnerability were involved, three years’ imprisonment had been imposed for mere attempts to commit the offence.

The Appellant denied administering the drugs to the Victims. She argued that the forensic investigations were insufficient to show that only she could have administered the drugs to the babies. The Report showed at best that she had access to those drugs that were found in her body in September 2016. On the other hand, many of the poisons were widely available over the counter, and a majority of them were not found inside her home. Furthermore, the police had been single-minded in investigating the Appellant to the exclusion of other suspects. They only belatedly conducted an incomplete screening of the medical records of the Victim’s family members, and did not conduct a search of their homes. In this regard, it was possible that BB1 had been administered the drugs by BB1’s paternal grandmother, who was a senior assistant nurse at Tan Tock Seng Hospital (“TTSH”) and disapproved of BB1’s babysitting arrangement with the Appellant. It was also possible that BB2 had been administered the drugs by either or both of BB2’s parents. They had sent BB2 to the hospital only in the evening despite collecting her in the morning.

The Appellant argued that a global sentence of not more than two years’ imprisonment was appropriate. Among other things, her contribution to society as a nurse was a mitigating factor and the poisons had no lasting side-effects on the Victims. She had also been suffering from various mental conditions including major depressive disorder, such that she was not an appropriate example to deter others from committing similar offences. Several of the cases highlighted by the Prosecution involved domestic workers poisoning their employers or employers’ families as revenge, but there was no such consideration in the present case, and those cases involved substances such as liquid soap and insecticide that were not meant for consumption under any circumstances.

Decision below

The Judge found that the strength of the circumstantial evidence was such that an irresistible inference could be drawn that the Appellant had administered the drugs to the Victims.

He observed that the drugs that were found in the blood of BB1 were connected to the Appellant primarily due to the timing of the symptoms and the fact that all the drugs found in the blood of BB1 could be traced to identical drugs that the Appellant had taken or been prescribed as medication. Furthermore, he was satisfied from the circumstances that no one other than the Appellant had administered the drugs found in BB1. He found that there was no evidential basis for the suggestion by the Defence that BB1 could have been poisoned by BB1’s paternal grandmother, whom he found an honest witness. There was also no suggestion by the Defence that M1 had poisoned BB1 and no evidence to indicate as much, and her actions was consistent with the behaviour of a concerned mother who sought medical attention for her daughter when something was wrong with her. The symptoms observed of BB1 being under the influence of medication or drugs which were found in her blood were right after BB1 was picked up from the Appellant’s home. It was also “most compelling” that all ten drugs found in the blood of BB1 coincided with the different drugs that the Appellant had consumed before and had access to.

As for BB2, her change in behaviour before and after the babysitting session with the Appellant could only be explained by the drugs found in her urine and blood samples. Since parties had agreed that Dr Looi and the Appellant’s two daughters did not administer the drugs to BB2, only the Appellant could have done so while BB2 was at her home. The Judge found no basis for the suggestion that either or both of BB2’s parents could have poisoned BB2. The Appellant also had recent access to those specific drugs that were found in BB2’s samples as well as the opportunity to administer them to her.

He additionally found the Appellant an unreliable witness who had sought at every turn to distance herself from the poisons found in the bodies of the Victims; yet, these attempts were futile in light of her medical records.

On sentencing, the Judge considered, inter alia, the young age of the Victims, the nature of the drugs, which put their lives at risk, and the high level of trust reposed in the Appellant by the Victims’ mothers. Deterrence and retribution were therefore relevant sentencing principles. He agreed with the Prosecution’s submissions that the applicable sentences were in the range of three to four years’ imprisonment per charge, based on the sentencing precedents tendered. He found that the sentence for each offence should be three and a half years’ imprisonment, with the totality principle not being highly relevant as the total sentence was not crushing or disproportionate to the Appellant’s criminality.

The arguments on appeal

On appeal, the Appellant raises essentially the same contentions as in the proceedings below. She argues that it was not shown that she even possessed all the drugs at the material time: only one drug was found in her home along with mere traces of two others. She also questions the Judge’s reliance on the Report: as it was addressed to an investigator from the Serious Sexual Crime Branch of the Criminal Investigation Department (“CID”), this is said to imply that she was a victim of a sexual assault, and may in fact have been drugged by her assailant. She argues that the admission of the Report was tantamount to a violation of her right against self-incrimination. She submits that this right is a constitutional right or, at the very least, a statutory one as recognised in s 22(1) of the Criminal Procedure Code (Cap 68, 2012 Rev Ed) (“CPC 2012”). She submits as well that it ought to have been excluded, being similar fact evidence that does not fall within the categories of ss 14 or 15 of the Evidence Act (Cap 97, 1997 Rev Ed) (“EA”) for the admission of such evidence. In any event, even if the Report is found to be prima facie admissible, it is contended that it should be excluded pursuant to the common law discretion to exclude voluntary statements where the prejudicial effect of the evidence exceeds its probative value. The Appellant argues that if the Report is so excluded, the Prosecution’s case must fail, as the remaining evidence does not justify a finding of her guilt beyond reasonable doubt.

Additionally, she submits that the Judge erred in relying on the timing of symptoms exhibited by the Victims to find that the drugs were administered to them while they were in the Appellant’s care. Amongst other things, BB1’s family members had testified that BB1 appeared sleepy, tired and “not her normal self” sometime in the end of November 2016, even though the Appellant had not babysat BB1 between 22 November and 4 December 2016. As for BB2, given that the doctors had testified that the effects of the drugs would be most pronounced upon administration of them, there was a strong inference that the drugs were administered to BB2 a short time prior to her admission to hospital, as opposed to earlier in the day when she was at the Appellant’s home. Finally, the Appellant reiterates that other individuals could have administered the drugs to the Victims, and contends that the failure of the police to conduct timely investigations into other potential suspects has caused the loss of vital evidence in determining her guilt or innocence . The Appellant does not raise any arguments as to the appropriate sentence in the event that her conviction is upheld.

The Prosecution argues that the Judge correctly found that the Appellant had access to all the drugs found in the Victims’ bodies. It was undisputed that she had access to these drugs in 2016, and the claim that she could have involuntarily consumed them was a “preposterous one” being raised for the first time on appeal. Indeed, she had agreed in cross-examination that she had consumed all the medication which was found in her body pursuant to the Report. Furthermore, it is submitted that the Report was properly obtained through the exercise of the police’s investigative powers. As the Appellant had expressly consented to the admission and use of the Report by the Prosecution at the trial below, it is argued that it is not open to her to retrospectively retract her consent now. In any case, the Prosecution submits that it was not only entitled to rely on the Report but “duty-bound” to do so, the Report being relevant material and, as the “fruits of investigations”, the “property of the community to ensure that justice is done”. The Prosecution further argues that the argument on the constitutional right against self-incrimination must fail as, inter alia, the Court of Appeal in Public Prosecutor v Mazlan bin Maidun [1992] 3 SLR(R) 968 (“Mazlan”) had held that there is no constitutional right against self-incrimination, as was also acknowledged by the Appellant. It argues as well that the Report does not contravene the rule against similar fact evidence, since it does not constitute past misconduct that is being used to prove that she has a propensity to commit offences of that sort, nor is it used to collaterally attack her character. Furthermore, it is submitted that the Report should not be excluded since its probative value outweighs its prejudicial effect, and its reliability is not in dispute. The Prosecution submits that even if the Report is excluded from consideration, the Appellant’s convictions are safe.

The Prosecution argues as well that it is unclear that the method through which the drugs was administered to the infants is material, since it was undisputed that the drugs had been unlawfully administered to the Victims, and the only issue in dispute was as to the identity of the perpetrator. It is argued that in any case, an analyst from the HSA had testified that the drugs could have been administered orally or injected. The Appellant had in fact informed the police that she crushed half a tablet of “Piriton” (chlorpheniramine), mixed it with water and fed it to BB1. The Prosecution submits that although the scientific evidence cannot precisely pinpoint the exact time at which the drugs were administered, that evidence together with the presentation of the Victims’ symptoms sufficiently enabled the Judge to correctly triangulate the time period in which this likely occurred. Finally, the contention that the police failed to properly investigate other potential suspects was untrue, as everyone who stayed with the Victims or had close access to them were screened to ascertain whether they had access to prescription drugs. The Prosecution observes that the Appellant has not set out particulars regarding any argument against the sentence imposed, and that in any event the said sentence was appropriate.

My decision on the appeal against conviction

The law

Section 328 of the Penal Code provides as follows:

In relation to the two charges faced by the Appellant, it must therefore be proved by the Prosecution that: (a) the Appellant administered the drugs to the Victims; (b) with intent to cause hurt to them. As to the latter element, the Prosecution had submitted below, and the Judge appeared to implicitly accept, that the requisite intention and knowledge could be judged and inferred from her objective conduct and all the surrounding circumstances (Muhammad Khalis bin Ramlee v Public Prosecutor [2018] 5 SLR 449 (“Muhammad Khalis”) at [42]). Where it is shown that a reasonable person in the position of an accused person would have known that a certain outcome would result from his acts, he would have to prove or explain how and why he did not in fact have such knowledge (Muhammad Khalis at [44]). As the Appellant does not dispute this element (her contention being the former element, viz, that she was not the one who administered the drugs to the Victims), I say no more on this, save to note that where the former element is satisfied, the objective circumstances would point to the requisite intention. This would be in light of her position as a nurse and her acceptance in cross-examination that medication taken without prescription or in an incorrect dosage could cause harm to infants.

Issues to be determined

The following issues therefore arise for consideration:

The Appellant’s access to all of the drugs

In my view, the Judge was justified in finding that the Appellant had access to all of the drugs which were found in the Victims’ bodies.

Propriety of the use of the Report

The investigating officer in the present case testified that the Report was obtained in the course of his investigations, as he discovered that the Appellant was involved in another case in which her blood and urine samples had been taken and sent to the HSA for analysis on 6 September 2016. He therefore obtained the Report which was dated 28 September 2016. The Appellant had stated in re-examination that it was addressed to the Serious Sexual Crime Branch of the CID, as she was having a quarrel with Dr Looi and thought she might have been pregnant. As noted above (at [3]), the Appellant argues that she had allowed these medical tests to take place as part of her cooperation with investigations.

The Appellant’s samples were therefore voluntarily given, as opposed to being mandated under such statutes as the Registration of Criminals Act 1949 (2020 Rev Ed), which provides that a body sample may be taken from arrested persons, convicted persons and prisoners for forensic DNA analysis; and it is, inter alia, an offence if they refuse to provide or obstruct the taking of a sample without reasonable excuse (ss 13B(1) and 13E(5) respectively). There is also apparently no statutory prohibition against the use of such evidence in subsequent judicial proceedings, in the absence of any regulations for the manner in which such samples should be taken: see, in contrast, the Misuse of Drugs (Urine Specimens and Urine Tests) Regulations (1999 Rev Ed) reg 6, which mandates a procedure for the taking and depositing of urine samples for tests under s 31 of the Misuse of Drugs Act 1973 (2020 Rev Ed).

Nonetheless, I had concerns whether the use of the Report was proper, since the Appellant had given the sample for a specific purpose, separate from the investigations into the current case, where she was the complainant of a possible offence. To my mind, there could be arguments why information obtained from her complaint should not be used against her: aside from the privilege against self-incrimination, it may be envisaged that privacy or confidentiality rights could be engaged, particularly as these related to information about her own body. It was best for parties to be invited to ventilate possible arguments.

Having considered the arguments made before me, I have concluded that the use of the Report is proper as:

(a) It is relevant and admissible as evidence;

(b) There is no bar to its use either on the basis of

(i) The privilege against self-incrimination; or

(ii) The rule against similar fact evidence; and

(c) There is nothing to attract the exercise of the Court’s discretion against its use in the proceedings.

In addition, I am satisfied that there is no separate right of privacy or confidentiality relating to a person’s body that would need to be vindicated in this context.

(1) Relevance and admissibility

I agree with the Prosecution that the Report is relevant and admissible under several of the general categories of relevant facts under the EA. Its admissibility is also supported by s 263 of the CPC 2012, which provides that the reports of certain qualified persons on any matter or thing duly submitted to them for analysis are admissible as evidence of the facts stated therein “in any criminal proceeding under this Code”.

The Prosecution argues that the fact of the Appellant’s access to the drugs in September 2016, when the Report was made, is a relevant fact under, amongst others, ss 7, 9 and 11 of the EA. Section 7 provides for the relevance of facts “which are the occasion, cause or effect, immediate or otherwise, of relevant facts or facts in issue, or which constitute the state of things under which they happened or which afforded an opportunity for their occurrence or transaction”. Section 9 provides for the relevance of facts which, inter alia, are “necessary to explain or introduce a fact in issue or a relevant fact, or which support or rebut an inference suggested by a fact in issue or a relevant fact”. Section 11 provides that facts which are not otherwise relevant are relevant if: (a) “they are inconsistent with any fact in issue or relevant fact”; or (b) “by themselves or in connection with other facts they make the existence or non-existence of any fact in issue or relevant fact highly probable or improbable”.

The Report, which is prima facie evidence that the subject drugs were detected in her blood and/or urine, speak to an effect of the relevant fact of her access to the drugs (s 7 of the EA). It also supports an inference suggested by another relevant fact of her having been prescribed some of these drugs, that inference being that she had access to and was familiar with them (s 9 of the EA). For that reason, it also goes towards making highly probable the fact that she did indeed have access to and was familiar with the drugs (s 11(b) of the EA). Furthermore, the Report contradicts the Appellant’s testimony that she did not have some of the drugs in her house or was not familiar with them (s 11(a) of the EA). For example, she had on the stand denied having (dextro)methorphan in her house, although she stated in a statement to the police that she had received it when previously admitted to TTSH; and denied being familiar with, amongst others, (pseudo)ephedrine.

The breadth of s 263 of the CPC 2012 under which the Report was produced also indicates that it is properly admissible in the present proceedings. Section 263 provides that:

…

The Report may be considered to have been made on matters “duly submitted” to the analyst for his consideration, in the sense of “properly”, ie, proof of identity of the article sent to the qualified person with the articles examined by him must be established (State of Orissa v Kaushalya Dei AIR 1965 Ori 38 at [8], in relation to the former s 510 of the Indian Code of Criminal Procedure 1898 (“Indian CPC 1898”); the present s 293 of the Indian Code of Criminal Procedure 1973 (“Indian CPC 1973”)). Significantly, the phrase “any criminal proceeding under this Code” does not appear to require that the subject report is only admitted in proceedings that it was sought in contemplation of. It has been held that the phrase does not refer to judicial proceedings only and that “if such a report is made in any kind of proceeding under the Code…that can be used as evidence in any enquiry, trial or other proceeding under the Code” (Abdul Rahiman v The State of Mysore (1972) CriLJ 406 at [9], on s 510 of the Indian CPC 1898). It is noted that s 510 of the Indian CPC 1898 was intended to save the then-small number of government experts, whose evidence was frequently needed in the criminal courts, the need of appearing in Court (Law Commission of India, 41st Report (The Code of Criminal Procedure, 1898), September 1969 at para 41.1).

As such, I agree with the Prosecution that the Report was “properly obtained through the exercise of the police’s investigative powers”. It was also properly admissible pursuant to the general categories of relevant facts in the EA discussed above as well as s 263 of the CPC 2012. It is also significant that in the present case, the maker of the Report did provide oral evidence and was cross-examined by the Defence on the Report. There therefore cannot be an objection by the Appellant that such examination was required by her but not provided, having regard to the language of s 263 of the CPC 2012.

As a matter of clarification, however, it may not be entirely correct for the Prosecution to argue that it was “duty bound” to rely on the Report in the present proceedings. It is true that the Prosecution has a “duty to the court and to the wider public to ensure that only the guilty are convicted, and that all relevant material is placed before the court to assist it in the determination of the truth” (Muhammad bin Kadar v Public Prosecutor [2011] 3 SLR 1205 (“Kadar”) at [200]; reiterated in Muhammad Nabill bin Mohd Fuad v Public Prosecutor [2020] 1 SLR 984 (“Nabill”) at [37]). This reflects the fact that the Prosecution acts “at all times in the public interest” (Nabill at [37]). However, it does not necessarily follow that the Report falls under the “fruits of investigations” which are to be disclosed to ensure that justice is done (Kadar at [200]). Rather, the notion of “fruits of investigations” in R v Stinchcombe [1991] 3 SCR 326 (“Stinchcombe”), which was cited by the Court of Appeal in Kadar (at [90] and [200]), was concerned with the Crown’s obligation to disclose to the Defence statements by a witness who had earlier given evidence at a preliminary inquiry that was favourable to the accused. It was in that context that the Supreme Court of Canada observed that “the fruits of the investigation which are in the possession of counsel for the Crown are not the property of the Crown for use in securing a conviction but the property of the public to be used to ensure that justice is done. In contrast, the [D]efence has no obligation to assist the [P]rosecution and is entitled to assume a purely adversarial role towards the [P]rosecution” (at [11]–[12]).

The subject matter of required disclosure by the Prosecution under Stinchcombe is therefore described as “fruits of the investigation”, since it likely includes (R v Gubbins [2018] 3 SCR 35 (“Gubbins”) at [22], citing R v Jackson (2015) ONCA 832 at [92]–[93]):

As stated by the Supreme Court of Canada in Gubbins, the “fruits of investigation” refers to the police’s investigative files, as opposed to operational records of background information, and is information “generated or acquired during or as a result of the specific investigation into the charges against the accused” (at [22]). As such, it is not immediately obvious that it should encompass information generated pursuant to a separate investigation commenced by an accused person.

(2) Privilege against self-incrimination

The Appellant also argues that the admission of the Report violates the privilege against self-incrimination. The privilege has been stated as expressing the rule that “no one is bound to answer any question if the answer thereto would, in the opinion of the judge, have a tendency to expose the deponent to any criminal charge, penalty or forfeiture which the judge regards as reasonably likely to be preferred or sued for” (Blunt v Park Lane Hotel [1942] 2 KB 253 at 257, cited in AT & T Istel v Tully [1993] 1 AC 45 at 67). It is part of the broader right to silence, which in turn “does not denote any single right, but rather refers to a disparate group of immunities” including, amongst others, a general immunity, possessed by all persons and bodies, from being compelled on pain of punishment to answer questions posed by other persons or bodies; or being compelled on pain of punishment to answer questions the answers to which may incriminate them (R v Director of Serious Frauds Office, ex p Smith [1993] AC 1 at 30–31). The privilege can cover “any piece of information or evidence on which the prosecution might wish to rely in establishing guilt”, including in deciding whether or not to prosecute a person (Den Norske Bank ASA v Antonatos and another [1999] QB 271 at 289). It is noted that privilege prevents the production of evidence and does not affect its admissibility in evidence once produced, which depends on its relevance (R v George Edward Tompkins (1977) 67 Cr App R 181 at 184; R v Governor of Pentonville Prison ex p Osman [1990] 1 WLR 277 at 310). What evidence is relevant is in turn defined by the EA (Basil Anthony Herman v Premier Security Co-operative Ltd and others [2010] 3 SLR 110 at [25]).

As noted by the parties, the Court of Appeal held in Mazlan that the right to silence is not a fundamental principle of natural justice which is included in the word “law” in Art 9(1) of the Constitution of the Republic of Singapore (1985 Rev Ed) (which in turn provides that “[n]o person shall be deprived of his life or personal liberty save in accordance with law”; at [15]). According to the Court, to recognise it as a constitutional right “would be to elevate an evidential rule to constitutional status”, even though it was not explicitly expressed in the Constitution (at [15]). There was “no specific constitutional or statutory provision protecting such a ‘right [to silence]’ or such a ‘privilege [against self-incrimination]’)” (at [13]). In fact, other provisions in the Criminal Procedure Code (Cap 68, 1985 Rev Ed) (“CPC 1985”) which expressly derogated from such a right, ie, ss 122(6) and 123(1) of the CPC 1985 (predecessors to ss 23(1) and 261(1) of the CPC 2012), had been upheld as valid and constitutional (at [18]–[19]). It is noted that the Court of Appeal appeared to treat the right to silence and the privilege against self-incrimination as co-extensive although, as discussed, they are not: the privilege protects the right of witnesses not to incriminate themselves, not their right to remain silent (see Queensland Law Reform Commission, The Abrogation of the Privilege Against Self-Incrimination (Report No 59) (December 2004) at para 1.4).

In the face of the clear words in Mazlan, it does not seem open to the Appellant to argue that the privilege against self-incrimination may constitute part of the fundamental rules of natural justice incorporated into ‘law’ for the purposes of Art 9(1) of the Constitution (ie, with the effect that any deprivations to life or personal liberty must be in accordance with it). This is notwithstanding the Privy Council’s earlier observation in Haw Tua Tau and others v Public Prosecutor [1981–1982] SLR(R) 133 that “what may be properly be regarded by lawyers as rules of natural justice change with the times” (at [26]) and the holding by the Court of Appeal in Yong Vui Kong v Public Prosecutor [2015] 2 SLR 1129 that the fundamental rules of natural justice are “procedural rights aimed at securing a fair trial” (and do not contain substantive legal rights) (at [64]). I also do not think the Appellant is correct to argue that the privilege against self-incrimination is, at least in Singapore, part of the fair hearing rule that was observed to be applicable in both the administrative and constitutional contexts in Yong Vui Kong v Attorney-General [2011] 2 SLR 1189 at [105]. As set out by Warren Khoo J in Stansfield Business International Pte Ltd v Minister for Manpower (formerly known as Minister for Labour) [1999] 2 SLR(R) 866 at [26], the fair hearing rule requires:

This is quite separate from the privilege against self-incrimination as described above, and the Appellant has not shown how the privilege can be a “bedrock” for the conduct of a fair hearing. Although the Grand Chamber of the European Court of Human Rights has held that the privilege lies “at the heart of the notion of a fair procedure” under Art 6 of the European Convention of Human Rights on the right to a fair trial despite it not being specifically mentioned therein (John Murray v United Kingdom (1996) 22 EHRR 29 at [45]; Bykov v Russia [2009] ECHR 441 at [92]), those observations were made in a different statutory context and cannot apply here in light of the decision in Mazlan.

Additionally, although s 22(2) of the CPC 2012 does statutorily express the privilege against self-incrimination, I agree with the Prosecution that it does not apply to the present case. Section 22(1) of the CPC 2012 provides that a police officer “may examine orally any person who appears to be acquainted with any of the facts and circumstances of the case”, while s 22(2) states that the person examined by the police is “bound to state truly what he knows about the facts and circumstances of the case, except that he need not say anything that might expose him to a criminal charge, penalty or forfeiture”. Section 22(2) of the CPC 2012 thus permits the person being questioned to maintain his silence on matters that may be personally incriminating (Lim Thian Lai v Public Prosecutor [2006] 1 SLR(R) 319 at [17]–[18]) and to that extent preserves the privilege against self-incrimination for that person (Michael Hor, “The Privilege against Self-Incrimination and Fairness to the Accused” (1993) SJLS 35 atp 38; Ho Hock Lai, “The Privilege Against Self-Incrimination and Right of Access to a Lawyer” (2013) 25 SAcLJ 826 at [6]).

However, s 22(2) of the CPC 2012 relates to the oral examination of a witness, as indicated by ss 22(1) and 22(3)–(7), the latter of which provisions contemplate the reduction of that examination into writing. This is also the case in India, where it has been held that statements reduced into writing pursuant to the equivalent of s 22(3) (s 161(3) of the Indian CPC 1973) refers to “all that is stated by a witness to a police officer or officers during the course of investigation” (Asan Tharayil Baby v State of Kerala (1981) CrLJ 1165 at [13]. On the other hand, the report of the Chemical Examiner on the examination of blood of an accused person did not appear to be the subject to the statutory limitations on such statements (see Ukla Kohle v State of Maharashtra (1963) AIR 1531). It therefore seems to me that s 22(2) of the CPC cannot be taken to represent a privilege of self-incrimination as regards the Report. This is also consistent with the fact that the common law privilege does not exclude real evidence, as I turn now to discuss.

The Appellant faces at least three difficulties in any assertion of the privilege against self-incrimination at common law in my view. First, the privilege does not appear to cover non-testimonial evidence such as her blood and urine samples and the consequent Report. Second and more fundamentally, the privilege does not cover instances where there was a lack of compulsion to provide the evidence in question. Third, it would seem that she has lost any such privilege by reason of not claiming it in the proceedings below.

On the first, the privilege does not cover real evidence such as her samples. In the UK, it was held that the privilege, being primarily concerned with respecting the will of an accused person to remain silent, does not extend to the use in criminal proceedings of material that may have been obtained compulsorily from the accused but which have an “existence independent of [his] will … such as, inter alia, documents acquired pursuant to a warrant, breath, blood and urine samples and bodily tissue for the purpose of DNA testing” (Attorney-General’s Reference (No 7 of 2000) [2001] WLR 1879 at [34], citing Saunders v United Kingdom (1996) 23 EHRR 313 at [69]; see also George v Coombe [1978] Crim LR 47 and R v Kearns [2002] 1 WLR 2815 at [53(4)]). This is similarly the case in Australia (Sorby and another v Commonwealth of Australia and others (1983) 46 ALR 237 at 243 to 244). As discussed by the Supreme Court of Canada in R v Stillman [1997] 1 SCR 607:

The Court therefore held that the privilege against self-incrimination was to be confined to testimonial evidence and not real evidence which, in that case, consisted of hair samples, buccal swabs and teeth impressions which were taken from the accused without consent while he was in custody. Such real evidence would instead generally fall to be treated under the search and seizure guarantee under s 8 of the Canadian Charter of Rights and Freedoms (“the Charter”) (which provides that “everyone has the right to be secure against unreasonable search or seizure”).

Such a constitutional right could potentially have been relevant if recognised here. The present case is somewhat akin to R v Cole [2012] 3 SCR 34, where school officials had voluntarily provided to the police an employee’s laptop containing child pornography, which had been obtained by the school technician in the context of routine maintenance activities. The Supreme Court of Canada held that the police infringed the accused’s rights under s 8 of the Charter, since the fact that the laptop had been lawfully acquired by the school board for its own administrative purposes did not vest in the police a delegated or derivative power to use it for a criminal investigation (at [67]). The majority was however of the view that the unconstitutionally obtained evidence should not be excluded under s 24(2) of the Charter as its admission would not bring the administration of justice into disrepute. A similar concern has been recently expressed in the United States: see, eg, “Police to stop using rape victims’ DNA to investigate crimes”, Associated Press (24 February 2022) and Azi Paybarah, “Victim’s Rape Kit Was Used to Identify Her as a Suspect in another Case”, The New York Times (15 February 2022), citing the constitutional protection against unreasonable searches and seizures by the government. However, given the absence of a similar right recognised here, the preferable approach to determining whether the Report should be excluded rests on the residual discretion at common law to exclude evidence where its prejudicial effect outweighs its probative value, as discussed below (at [68]).

Second, I agree with the Prosecution that the Appellant faces the further difficulty that she was not compelled to provide her blood and urine samples for analysis, or agree to the admission of the Report into evidence. This follows from the above definition of the privilege in [50] and [56], ie, the privilege not to be compelled to give evidence against oneself, or nemo tenetur prodere seipsum (‘no one is bound to betray oneself’) (RH Helmholz, “Origins of the Privilege Against Self-Incrimination: The Role of the European Ius Commune” (1990) 65 New York University Law Review 962 at p 962). It is also confirmed by the statutory preservation of the principle in various jurisdictions. For example, s 10(1) of the Evidence Act 1977 (Qld) provides that “nothing in this Act shall render any person compellable to answer any question tending to criminate the person”. In relation to civil proceedings in the UK, s 14(1) of the Civil Evidence Act 1968 (UK) provides for the “right of a person … to refuse to answer any question or produce any document or thing if to do so would tend to expose that person to proceedings for an offence or for the recovery of a penalty”. Furthermore, s 60 of the Evidence Act 2006 (NZ) provides for the application of the privilege where a person is “required to provide specific information” and that information would be likely to incriminate him for an offence, “self-incrimination” being defined in s 4 as “the provision by a person of information that could reasonably lead to, or increase the likelihood of, the prosecution of that person for a criminal offence”.

Third, it seems that by not previously asserting any such privilege, it has been lost. As stated by Lindsay J in O Ltd v Z [2005] EWHC 238 (Ch) (“O Ltd”) (at [58]):

Thus, in O Ltd, where the defendant had provided computer and other recorded material to a computer expert engaged by the claimant, pursuant to a search order on which paedophile pornography of a serious nature was found, the expert invited the Court to grant permission for those materials to be handed to the police. Although the defendant later indicated that he wished to assert the privilege against self-incrimination, Lindsay J held that the privilege had been lost on an objective view of his behaviour; he had handed the offensive material to the expert without claiming the privilege (at [70]). Similarly, in R v Clyne [1985] 2 NSWLR 740, where the Appellant had made admissions in his oral evidence in proceedings where he had disputed a bankruptcy notice, the New South Wales Court of Criminal Appeal held that he could not challenge the admission of such evidence when he was subsequently charged with giving a false statement to an officer of the Reserve Bank of Australia under certain banking regulations. The Court observed (at p 747), citing the Privy Council decision in R v Coote (1873) LR 4 PC 599, that evidence given by a witness is ordinarily admissible even if it may be incriminating, unless he has already claimed the privilege of self-incrimination and been denied it by the presiding Judge. It added that the mere fact that a witness on the stand is “ordinarily obliged to answer questions does not of itself make those answers involuntary in the sense that it is relevant if those answers are later sought to be used against the witness in criminal proceedings” (at 747).

Given therefore that counsel for the Appellant had not previously objected to the admission of the Report, it does not seem open to her to object to its production by way of asserting privilege now.

Furthermore, in support of her argument that the Report “unfairly prejudiced and incriminated” her in the present proceedings, the Appellant relies on Wong Kim Poh v Public Prosecutor [1992] 1 SLR(R) 13 (“Wong Kim Poh”) for the proposition that “as a general rule, the prosecution is not at liberty to rely on prejudicial evidence merely because it [was] included in the accused person’s statement” (Jeffrey Pinsler SC, Evidence and the Litigation Process (LexisNexis, February 2020) (“Pinsler, Evidence”) at para 3.053). However, I agree with the Prosecution that Wong Kim Poh does not assist her. There, the accused who had been charged with murder had made several statements under s 121 of the CPC 1985 which were admitted under s 122(5). His confession that he had stabbed the deceased was contained in only the last paragraph of the statements, while the 16 preceding paragraphs showed that prior to the killing, he was a pimp and had lived on the immoral earnings of his girlfriend. After holding that, inter alia, there was no basis to suggest that the trial judges had been adversely influenced by these statements concerning the character of the accused, Yong CJ opined that the preceding paragraphs of the statements (at [15]):

This is however not the case at hand, given that the matters in the Report do not have “nothing whatsoever to do” with the subject offences but instead are specifically connected with the facts in issue, as discussed above.

(3) Similar fact evidence

The Appellant additionally argues that the Report constitutes similar fact evidence that does not does not fall under ss 14 and 15 of the EA, which are the governing provisions for the admissibility of such evidence. However, the Prosecution argues that the rules governing similar fact evidence are not engaged. This is as it has not argued or suggested that the Appellant’s consumption of the medication amounted to misconduct or an offence, or that such consumption meant that she was disposed to drug babies.

I agree with the Prosecution that the Report does not fall foul of the rule against similar fact evidence. Such evidence relates to circumstances in which an accused person acts on occasions other than the one which gave rise to the charged offence, which is sought to be admitted to prove the guilt of the accused (Pinsler, Evidence at para 3.001). It consists of conduct that is not specifically connected with the facts in issue (which falls under ss 6–10 of the EA, being illustrations of “different instances of the connection between cause and effect which occur most frequently in judicial proceedings”), but which is merely similar in nature to those facts in issue (Pinsler, Evidence at paras 3.004–3.005, citing JF Stephen, The Indian Evidence Act, with an Introduction on the Principles of Judicial Evidence (Thacker, Spink & Co, 1872) at p 55). The rule against similar fact evidence is therefore concerned with the use of such evidence in an impermissible way, ie, to infer from a person’s past crimes that he has a propensity or tendency to commit similar crimes (Tan Meng Jee v Public Prosecutor [1996] 2 SLR(R) 178 (“Tan Meng Jee”) at [41]; Public Prosecutor v Ranjit Singh Gill Menjeet Singh and another [2017] 3 SLR 66 at [17]). In other words, the purpose for which the evidence is sought to be admitted is crucial (Michael Anak Garing v Public Prosecutor and another appeal [2017] 1 SLR 748 at [8]; Tan Meng Jee at [37]).

It follows from the above analysis on ss 7 and 9 of the EA that the Report is in fact specifically connected with the facts in issue vis-à-vis the subject offences and so is not similar fact evidence. The proximity of about two to three months between the time the samples in the Report were collected and the time of the subject offences also does not present an issue to the said relevance. As noted in SC Sarkar, The Law of Evidence in India, Pakistan, Bangladesh, Burma, Ceylon, Malaysia and Singapore (LexisNexis, 18th Ed, 2014), s 7 of the EA is expressed in “terms much wider than those of s 6 which makes relevant facts forming part of the same transaction … facts though not strictly forming part of the transaction may be so closely connected with it that they tend to prove or disprove or explain the transaction under enquiry” (at p 300) [emphasis in original]. The prima facie evidence that the drugs were found in her blood and urine prior to the subject offences are not strictly connected with her having possibly administered the drugs to the babies, but nevertheless have in common the relevant fact of her access to the drugs towards the end of 2016.

I note as well that in Evidence, Professor Pinsler had, in discussing similar fact evidence, observed that while the rules of admissibility appear to apply regardless of whether an accused person had intentionally or unintentionally disclosed evidence of his bad character, it is “arguable that the [P]rosecution should not be entitled to take advantage of the accused’s error by adducing evidence which would otherwise be inadmissible” (at para 3.051). He observes that an important question is then whether the court has a discretion to exclude or limit the use of such evidence, which I turn to consider next.

(4) The court’s exclusionary discretion

The Appellant also argues that the Report should be excluded on the basis of the court’s residual discretion at common law to exclude evidence where its prejudicial effect outweighs its probative value (Kadar at [51]–[53]; Sulaiman bin Jumari v Public Prosecutor [2021] 1 SLR 557 (“Sulaiman”) at [44]). Probative value has been defined as its ability to prove a fact in issue or a relevant fact; while its prejudicial effect refers to how its admission may be unfair to the accused person as a matter of process (Sulaiman at [47]). Probative value also varies depending on the type of evidence in question and the context in which it is sought to be adduced: for example, where there is lack of evidence on a relevant issue, the probative value of the evidence is likely to be higher (Australian Law Reform Commission, Uniform Evidence Law Report (ALRC Report 102, 2005) at para 16.15). In Sulaiman, the Court of Appeal expressed the view that the exclusionary discretion is concerned with the reliability of the evidence in question, having regard to whether there was “some form of unfairness in terms of the circumstances and process by which they were obtained” (at [45]).

In my view, given there is no issue of unlawfulness in procuring the evidence or of bad faith, any prejudicial effect to the Appellant (as defined in [68]) is somewhat limited. On the former, the decision of Kadar demonstrates that it is “only serious irregularities, meaning those that materially affect the evidential value of a voluntary statement”, that will suffice for the court to exercise the exclusionary discretion (at [65]). A failure to caution a person under s 122(6) of the CPC 1985 or to be informed of rights under s 121(2) of the same would not in themselves suffice (at [66]). There, the Court of Appeal found that breaches of s 121 of the CPC 1985 and the relevant Police General Orders were “serious enough to compromise in a material way” the reliability of the certain statements (at [146]). In contrast, there are no such questions of breaches of procedure compromising reliability in relation to the Report.

On the latter, it has been held in England that bad faith may be a factor leading to the exclusion of evidence under s 78 of the Police and Criminal Evidence Act 1984 (UK) (“PACE”), which confers on the court the discretion to exclude prejudicial evidence on which the Prosecution proposes to rely. Furthermore, evidence may nevertheless be excluded for serious breaches of procedure even if the police have acted in good faith (Colin Carlton Alladice (1988) 87 Cr App R 380 at 386). For example, in Matto v Wolverhampton Crown Court [1987] RTR 337, the English High Court held that given the Crown Court’s finding that the police officers knew they were in excess of their powers and therefore acted in bad faith in administering a breathalyser test, circumstances existed for the exclusion of evidence under s 78 of PACE. Again, I do not view the police or the Prosecution as having acted in bad faith in the circumstances.

As for instances where body samples taken in one inquiry are used in another, it seems that in the absence of any regulations governing their use or retention, there might only be issues of unfairness which may justify excluding the evidence if the accused was assured that this would not take place. This was the case in R v Nathaniel [1995] 2 Cr App R 565, where DNA evidence was given by an accused person in relation to a possible offence of rape involving two Danish girls of which he was later acquitted, and the police failed to destroy his DNA profile as required under s 64 of PACE even though they had informed him, before taking the sample, that such destruction would take place were he cleared of the offence. The police instead entered it on the Metropolitan computer index which matched him to another offence, for which he was arrested and convicted. The English Court of Appeal observed that not only had there been a breach of s 64, but the accused “had in effect been misled in consenting to give the blood sample by statements and promises which were not honoured” (at 571). Subsequently, in Attorney General’s Reference (No 3 of 1999) [2001] 2 AC 91, concerning a DNA profile from a saliva sample from a defendant’s charge of burglary (for which he was acquitted) which later linked him to a rape, the House of Lords held that wrongfully retained samples would nevertheless be admissible subject to the exclusionary discretion under s 78 of PACE. Lord Steyn expressed the view that there ought to be “fairness to all sides”, ie, the accused, the victim and his or her family, and the public; and it was “in the interests of everyone that serious crime be effectively investigated and prosecuted” (at p 118).

Although s 78 of PACE casts the exclusionary discretion in wider terms than the common law discretion recognised in R v Sang [1980] 1 AC 402 (R v Stephen Cooke [1995] 1 Cr App R 318 at 328) and in Singapore (Kadar at [51]–[53]), as it permits the exclusion of evidence having an “adverse effect on the fairness of the proceedings”, the present case does seem to fall outside the ambit of a breach of procedure or impropriety such that the admission of the Report into evidence may be unfair to the Appellant as a matter of process.

This does not mean, however, that it would always be permissible for evidence voluntarily provided by a complainant to the police to be used subsequently against him in an unrelated criminal proceeding. It seems to me that this should be subject at least to the constraint that the evidence should have been reasonably obtained in respect of the initial suspected offence. Thus, where a complainant assists in investigations by, for example, handing over his phone, it would be reasonable for him to expect that the phone would be examined to the extent necessary for the offence complained of, and the evidence obtained from that investigation could be subsequently used against him in an unrelated criminal proceeding. That limitation does not arise here, since the Appellant’s blood and urine samples were obtained in relation to a suspected sexual offence against her, and would have been expected to be material evidence in relation to the alleged offence. Furthermore, it could certainly be the case that the court’s exclusionary discretion may still operate, where such evidence has more prejudicial effect than probative value. Again, however, that does not arise here since the Report was, as discussed, highly probative, with limited prejudicial effect.

(5) A right to privacy or confidentiality relating to information from one’s body

I was concerned that there should be some privilege or right recognised protecting confidentiality or privacy concerning information from one’s own body. I have some sympathy for the concern that there should be some limits on the use of information relating to one’s own body. That is why there are often statutory controls imposed on the taking of samples, such as the taking of blood, urine or alcohol samples. It is intrusive. Furthermore, the use of such intrusive information voluntarily given by a person for another purpose would on the face of it raise, at least at first blush, concerns about fairness and breach of confidence. There may be an expectation that disclosure is to be used only for the purpose the information was given for. Whether that expectation should be protected or vindicated is, of course, another matter.

On the other hand, there is, of course, no express statement of any such right, either in the Constitution or the relevant criminal law statutes. There will, to my mind, be substantial factors militating against the recognition of any common law right or privilege, especially given the position I have reached above in respect of the court’s discretion to exclude. At the same time, while it is well established that personal or private information is protected by the law of confidence (see, eg, X v Y and others [1988] 2 All ER 648), such information is nevertheless not protected where there is a reasonable suspicion that it relates to crimes, frauds or misdeeds, or misconduct of such a nature that ought in the public interest be disclosed to others (Malone v Metropolitan Police Commissioner [1979] Ch 344 at 377; Initial Services Ltd v Putterill and another [1968] 1 QB 396 at 405). Even if the related tort of misuse of private information were to be recognised here (see, eg, ZXC v Bloomberg LP [2022] 2 WLR 424, where the UK Supreme Court found that a person under criminal investigation who had not been charged with a criminal offence had a reasonable expectation of privacy in respect of information relating to that investigation), it is recognised that the prevention and detection of crime and fraud would be a countervailing public interest (Law Reform Committee, Singapore Academy of Law, Report on Civil Liability for Misuse of Private Information (December 2020) at para 2.26); and the latter would likely prevail in the circumstances. Additionally, the absence of even any nascent protection for the privilege against self-incrimination in our jurisprudence is a strong pointer against the development of such common law rights or protection. And for many of the same reasons that the court’s exclusionary discretion should not be exercised against the admission of the Report, I do not see that there is any scope for the development of any such right or privilege.

Effect on the case as a whole if the Report were excluded from consideration

I am satisfied that, in any event, any exclusion of the Report from consideration would not affect the correctness of the Appellant’s conviction.

This is because apart from the Report, a medical report from Everhealth Medical Centre showed that the Appellant had prescriptions for zolpidem (“Stilnox”), alprazolam (“Xanax”) and diazepam at the time of the offences. As mentioned at [16] above, an empty slab of zolpidem (“Stilnox”), a slab of chlorpheniramine (“Piriton”), a handkerchief with traces of diazepam and milk bottle with traces of zolpidem were also found in her home. In her fourth statement to the police dated 30 July 2018, she had furthermore explained why she had (dextro)methorphan, (pseudo)ephedrine, orphenadrine, as well as zolpidem, chlorpheniramine, alprazolam and diazepam. She ventured suggestions that while she could not remember how she was prescribed with triprolidine and ephedrine, they may have been prescribed to her while she was in Changi General Hospital. These account for all the substances that were found in the Victims and that are subject of the Charges (including the metabolites of the said drugs). Dr Looi also stated that in 2016, he must have brought chlorpheniramine and orphenadrine to the Appellant’s home. At the same time, parties agreed that (pseudo)ephedrine and/or ephedrine, (dextro)methorphan and triprolidine are available over the counter. Evidence apart from the Report therefore amply supports the Judge’s finding that the Appellant had access to the drugs.

Timing of the symptoms exhibited by the Victims

The Appellant argues that the two analysts from the HSA who were involved in preparing the toxicology reports in respect of the Victims were unable to say, based on those results, when the drugs were administered to the Victims. The Appellant submits that BB1’s family members had also testified that BB1 was drowsy even when the Appellant did not babysit her (as noted at [30] above). Furthermore, she argues that Dr Low Eu Hong (“Dr Low”), who had attended to BB1 when she was admitted to PEH on 9 December 2016, had found that the symptoms exhibited by BB1 that he observed were not due to drug poisoning, but hypoglycaemia; and that the Judge erred in not considering this fact. Additionally, Dr Tan and Dr Koh had testified that certain of the drugs, such as diazepam, would have an observable effect within minutes of being administered, with Dr Koh testifying that the effects would generally be the most pronounced at the start of the administration. All this points away from the Appellant having administered the drugs to the Victims, particularly given the reluctance of M1 to bring BB1 to the hospital on 9 December 2016, and the gap in time of over ten hours between the time F2 collected BB2 and the time she was sent to the hospital.

However, I agree with the Prosecution that while the scientific evidence could not indicate the precise time at which the drugs were administered to the Victims, the evidence of the Victims’ symptoms as observed by the Victims’ family members as well as the doctors who examined them supported the Judge’s finding that these symptoms were attributable to them being drugged while in the Appellant’s care.

In relation to BB1, M1 had testified that upon picking her up sometime between 6pm and 6.30pm on 9 December 2016, she observed that BB1’s eyes were swollen and she was “really drowsy” and “not really responding”. BB1’s maternal grandmother, who saw BB1 at around 7pm that day, similarly observed that she was “not her usual … self” and her head was moving from side to side. Dr Low, who subsequently saw BB1 at about 8pm that day, testified that BB1 was “very drowsy, very dazed and very floppy”, and it struck him that BB1 had been intoxicated by alcohol or sedatives from certain drugs.

The Appellant’s assertion of an error on the Judge’s part of insufficiently considering Dr Low’s evidence is also unfounded. Dr Low had stated in the medical memo (at [10] above) that there was “drug overdose/toxicity to alprazolam, dextromethorphan, pseudoephedrine and zolpidem” as well as “hypoglycema”. He testified that the former diagnosis was based on the confirmation that BB1 had toxicity to all the four drugs mentioned, and that the latter diagnosis was likely due to the first, in that since BB1 had been very sleepy and drowsy, she probably could not be fed properly and therefore had low blood sugar. There was therefore no error by the Judge in his appreciation of Dr Low’s evidence: the upshot of his evidence was that the two conditions were likely linked and based on an administration of drugs to BB1.

As for BB2, F2 had testified that she was sleeping when he picked her up between 6.30am and 7.20am that day. BB2’s parents testified that upon waking up while at home, BB2 was “very drowsy” and could not walk or stand. This was the case over the course of the day, and she was eventually brought to KKH around 5pm. A medical report from KKH stated that according to her parents, BB2 had rolled off her floor mattress onto the ground three to four times and was drowsy even after waking up, and not wanting to feed or drink as usual. The Appellant argues that the Judge did not appreciate however that the same medical report provides that BB2’s parents were not too concerned initially after picking her up from the Appellant’s home, “as she was still able to play and interact normally”. This, she argues, suggests that the severity of BB2’s symptoms were not as serious as attested to by BB2’s parents. Yet, that account does not entirely detract from the thrust of BB2’s symptoms after she was picked up from the Appellant’s home, according to her parents (as recorded in the medical report) and at trial. That medical report states that BB2’s parents brought her to KKH “in view of her drowsiness and lethargy”.

Furthermore, BB2’s symptoms when examined at KKH (at [13] above) were consistent with the drugs eventually found in her body. Dr Koh opined that an acute overdose of benzodiazepines (or sleeping pills) such as alprazolam, diazepam, oxazepam and temazepam could result in somnolence, impaired coordination, confusion and diminished reflexes, which were seen in BB2; and the main adverse reactions in an acute overdose of chlorpheniramine, an anti-histamine, included drowsiness and fatigue, which was also seen in BB2. She noted, however, that the main adverse reactions from an acute overdose of ephedrine were not seen in BB2 although, as noted at [14] above, that substance was detected in her urine sample which was taken from her two days after she was admitted to KKH.

The timing at which the Victims were said to have exhibited these symptoms are also consistent with the evidence of Dr Low, who stated that some of the drugs could be short acting and others long acting, but that the effects would by and large persist for six to eight hours, or even longer, depending on the dosage. Dr Tan and Dr Koh similarly testified that the various drugs would have different effects as regards their onset and duration of action. Accordingly, I am satisfied that the Judge did not err in finding that the timing of the symptoms exhibited by the Victims supported his conclusion that the Appellant had fed the drugs to the Victims.

Consideration of other possible suspects

In my view, the Judge did not err in finding that it was the Appellant, and not other suspects, who had poisoned the Victims. In relation to BB1, the Judge had considered that possible suspects other than the Appellant were the mother, grandmothers and the paternal aunt of BB1. However, the maternal grandmother and paternal aunt of BB1 testified that they did not feed BB1 any drugs in 2016. Their evidence was not challenged in cross-examination by the Defence. Furthermore, as mentioned above (at [22]), the Judge considered the evidence of the paternal grandmother of BB1, who testified that as a senior assistant nurse, she did not give patients their medication, as only registered nurses were allowed to do so. She stated that she also did not have access to medication in TTSH, where she worked, and was familiar with only some of the drugs that were found in BB1’s body. She testified that she had not administered such drugs to BB1, as any medication would be administered to BB1 by M1. At the same time, the Defence did not suggest at trial or on appeal that M1 could have administered the drugs to her. In my view, there was no basis to depart from the Judge’s findings that these other possible suspects had not administered the drugs to BB1.

In relation to BB2, the Appellant submits that the Judge erred in dismissing as “fanciful” her suggestion that BB2’s parents could have administered the drugs to BB2. She argued that F2 had a prescription for diazepam on 20 December 2016, although he had testified that there was no medication in the house at the material time. Furthermore, both parents had been awake the previous night, having been at a countdown party, and it was possible that they had fed BB2 diazepam so they could get some rest. However, apart from the fact that this was not put to either witnesses in court, investigations did not show that F2 or M2 were prescribed with any of the other drugs (save for diazepam in respect of F2) which were found in BB2’s body at the time. The Judge therefore did not err in considering that any evidence pointing towards them as the perpetrators was insufficient to found any reasonable doubt that the Appellant had not committed the offences.

Additionally, I disagree with the Appellant’s contention that she was strongly prejudiced as a result of the focus of investigations allegedly being on herself as the primary suspect, and that the Judge erred in failing to consider this. The police did not search any other homes than that of the Appellant, and had considered her the only suspect initially. However, this was based on the information received. F2 had called the police on 27 December 2016 at about 9.48pm, reporting that his daughter had been sent to a nanny, was drowsy when he had collected her, and had been admitted to hospital. He also reported that the nanny had slammed the door when he went to speak with her. As part of following up on the case, the police went to the Appellant’s house on 29 December 2016, whereupon the items mentioned above (at [16]) were seized. The police also interviewed other witnesses relevant to the investigation, including the parents of BB2, as well as Dr Looi and the Appellant’s two daughters. The investigating officer of the case from June 2018 also conducted screenings of the family members of the Victims to ascertain whether they were prescribed any drugs in the subject offences, in the period from September to December 2016. He therefore made inquiries with all the government hospitals, polyclinics, as well as private clinics that the family members stated they went to. The Appellant presently argues that the family members of the Victims could have somehow obtained these drugs from other sources without informing the police, and that it is prejudicial to her case that none of the Victims’ homes were searched in a timely manner. However, while the former suggestion is remotely possible, I consider that the Judge was justified in nevertheless drawing an irresistible inference that the Appellant committed the offences partly based on the evidence that was found concerning her, and the numerous inconsistencies surrounding her accounts for that evidence.

In other words, although the Appellant has not contested this, it is clear that the Judge was fully justified in assessing her credibility as lacking, and rejecting her evidence that she had not administered the drugs that were found in the Victims. Amongst other inconsistencies, her prescription records from Everhealth Medical Centre and her evidence on the stand regarding her familiarity with the drugs contradicted her Case for the Defence, where she stated that of the drugs found in the babies, she was familiar with alprazolam and zolpidem, but her prescription for the former only started in 2018 and she had never been prescribed or had taken diazepam from 2016. Her oral testimony also contradicted her statements to the police. For example, she had on the stand denied having (dextro)methorphan in her home, even though she addressed why she had them in her fourth statement to the police. She also testified that she was not familiar with (pseudo)ephedrine but had stated in that police statement that this had been prescribed to her as a medication for sore throat. Furthermore, as observed by the Judge, she had stated on two occasions to the police that she had been instructed by M1 to give BB1 “Piriton”, and could only say, when questioned on the inconsistency with her oral evidence that she had not given them any medication, that she had been “provoked by the police”. I also agree with the Judge that her explanation in cross-examination for the milk bottle with traces of zolpidem found in her home – that she had placed her own medicine in there as she urgently needed to go to the toilet – was “incredible” and an “afterthought”. It was, as he noted, a different explanation from that earlier provided in her examination-in-chief, which was that on the last occasion on which she babysat BB1, she had thought that BB1’s milk bottle was missing. She then asked her daughter to buy one, but later discovered BB1’s milk bottle in her stroller. As she did not want to upset her daughter, she poured the contents of BB1’s milk bottle (containing milk prepared by M1) into the new bottle.

Assessment of the evidence against the Appellant

The incontrovertible and undisputed evidence was that the babies in question were indeed poisoned, that is, they had drugs or foreign substances in their body. There was no question either that these substances were not meant to be there: none of these were of the sort that would be administered normally to such babies, nor would they have ingested them on their own. It must thus be inferred that someone must have made them ingest the drugs or poison.

With that as the starting point, the factual assessment must then turn to the determination of who had the opportunity and means to do so. The motive of the Appellant in doing so is not disclosed by the evidence, but that does not raise any reasonable doubt. The absence of a definitive motive does not undermine the strength of the evidence against her: many reasons or possibly no real reason could have underlain what she did. What her objective was in doing so would be known truly only to her. As examined above, no one else would reasonably have been able to commit the acts. The contrary version, that someone else could have committed the acts, was neither realistic nor reasonable. In particular, the access to drugs on the part of other possible suspects was not anywhere close to the same degree as the Appellant’s. While all of this was circumstantial, the cumulative effect of the evidence against her meant that there was an irresistible inference that she had indeed committed the offences with which she was charged. To recapitulate, the Appellant had:

(a) The opportunity to commit the offences, while the babies were in her care;

(b) She had access to all of the drugs, found in the babies, since as shown by the Report, she had consumed the same medication herself. Furthermore, several of the drugs or items with traces of the same were found in her home around the time of the offences;

(c) No one else would have had the opportunity or access to all the drugs to cause the babies to ingest them; and

(d) The babies did ingest the drugs, and no other explanation is plausible except that someone else caused them to do so.

Furthermore, as noted above, even without the Report, there was sufficient evidence to lead to the irresistible inference that she committed the offences. The burden on the prosecution was to prove beyond a reasonable doubt, not beyond all doubt or any shadow of a doubt.

In sum, I am satisfied that the Judge did not err in convicting the Appellant on the Charges.

My decision on the appeal against sentence

While the Appellant did not cite any ground for her appeal against sentence, I do not, in any event, see any reason to disturb the sentences meted out by the Judge. The offence under s 328 of the Penal Code is a serious one which carries a mandatory imprisonment sentence which may extend to ten years (Public Prosecutor v Tan Kok Leong [2016] SGDC 327 at [55]; Public Prosecutor v Ng Bee Ling Lana [1992] 1 SLR(R) 448 at [15]). The Judge had justifiably taken into account factors such as the young age of the Victims, which meant that they were wholly dependent on the Appellant while in her care; and the nature of the drugs administered, which placed the lives of the Victims at risk. For example, the evidence of Dr Low was that he would not have prescribed the subject drugs to children under six months of age, and would only administer small doses for those from six months to two years of age. Dr Koh furthermore testified that benzodiazepines would not be safe to consume without a prescription from a doctor, due to possible side effects. Dr Tan testified that the general concern with administering sedating medication to children under one year old was that this could affect their breathing. Multiple sedating agents would generally not be prescribed at the same time because of this as well as potential blood pressure and heart rate problems.

A number of other reasons point to a sufficiently heavy sentence. The Appellant was in a position of trust, with the parents having placed their young children in her care, expecting her to look after their safety and health. Instead, she betrayed the parents and those in her care, and endangered the health of the babies. Her culpability was substantial. In comparison, nothing weighty was raised in the mitigation below.

The sentence of three and a half years’ imprisonment in relation to each offence is also in line with that imposed in cases of a s 328 offence involving vulnerable victims (see, eg, Public Prosecutor v Titin Agustiana DAC 43422/09 (“Titin Agustiana”) (three years’ imprisonment) and Public Prosecutor v Dewi Supriyatin DAC 31293/05 (four years’ imprisonment), cited in Public Prosecutor v Fong Quay Sim [2010] SGDC 224 at [10]; and Public Prosecutor v Annisa DAC 932279/2014). I am not convinced that the distinction previously drawn by the Appellant between those cases and the present case – that they largely involved acts of domestic workers taking revenge on their employers or employers’ families, and substances which should not be consumed under any circumstances – would have been sufficient to warrant departing from these sentencing precedents. The domestic workers in these cases had pleaded guilty and would have been given a sentencing discount on that basis. Furthermore, the two-month-old twin victims in Titin Agustiana, to whom the accused had attempted to feed milk powder mixed with laundry detergent, did not actually ingest the tainted powder. As against this, the medical evidence was that the drugs which were administered by the Appellant to the Victims could have endangered their lives, as discussed above. There is accordingly no reason to disturb the sentence on any basis that it was, inter alia, wrong in principle or manifestly excessive (Public Prosecutor v UI [2008] 4 SLR(R) 500 at [12]).

For the reasons set out above, I dismiss the appeal in its entirety.