Zia Nabi analyses the concept of vulnerability as it has applied in homelessness cases since its introduction to statute in the 1970s and looks at whether recent judgments might be cause for optimism.
A person who is ‘vulnerable as a result of old age, mental illness or handicap or physical disability or other special reason, or with whom such a person resides or might reasonably be expected to reside’ has a priority need for accommodation (Housing Act (HA) 1996 s189(1)(c)). ‘Other special reason’ permits a cumulative assessment of characteristics, which taken together may render a person vulnerable for the purposes of the statutory test, even though they would not do so taken individually.
What is meant by ‘vulnerable’ is not statutorily defined. It is a mixed question of fact and law. In Crossley v City of Westminster  EWCA Civ 140
, Sedley LJ stated that given the evaluative nature of the exercise, judgements in priority need cases had to be made with especially careful regard for the statutory criteria and purposes and conscientious attention to the evidence. A decision-maker had an obligation to acknowledge, take into account and evaluate all relevant factors and consider with great care whether there were factors that rendered a person vulnerable for some other special reason.
Vulnerable – a very brief history
The term ‘vulnerable’ in the context of the modern homelessness statutory scheme first appeared in the Housing (Homeless Persons) Act (H(HP)A) 1977. That Act was then consolidated into HA 1985 Part III, which was itself replaced by the HA 1996. The test of what it means to be vulnerable remained unchanged.
The first significant case in which the term ‘vulnerable’ was considered was R v Waveney DC ex p Bowers  QB 238. Mr Bowers was a single man aged 59 who had suffered a brain injury and had alcohol issues. The Court of Appeal allowed his appeal. Vulnerability was described as being ‘less able to fend for oneself so that injury or detriment will result when a less vulnerable man will be able to cope without harmful effect’ (emphasis added).
The editors of the Housing Law Reports, rather optimistically as it turned out, stated:
This is the first case on vulnerability under the Act. The Court of Appeal have clearly adopted a low test of vulnerability, ie one that is not difficult to fulfil ((1981-82) 4 HLR 119).
The next case of note was R v Camden LBC ex p Pereira (1999) 31 HLR 317
, in which it was held that, when considering vulnerability:
… the council must ask itself whether Mr Pereira is, when homeless, less able to fend for himself than an ordinary homeless person so that injury or detriment to him will result when a less vulnerable man would be able to cope without harmful effects (emphasis added).
Although the Court of Appeal stated subsequently that the Pereira
reasoning should not be treated as the statutory test,1See, for example, Osmani v Camden LBC  EWCA Civ 1706 at para 38.
the reality was that it came to be treated as such, so much so that the vulnerability test became known by way of shorthand as the Pereira
test. The vulnerability test became increasingly difficult to satisfy, with the Court of Appeal generally deferring to the housing authority’s perceived expertise.
It will be seen that the comparator had in Pereira
, perhaps unintentionally, changed from the ‘less vulnerable man’ in Bowers
to the ‘ordinary homeless person’. Housing authorities successfully argued that accordingly the correct comparator was the ordinary (street) homeless person and not the ordinary person who was homeless. In Johnson v Solihull MBC  EWCA Civ 752
; July/August 2013 Legal Action
22, Arden LJ stated:
[T]he question of who is an ordinary homeless person and what characteristics they have is a question to be assessed in the real world. It is sadly not surprising that many homeless persons have drug issues, or that many homelessness services are involved with dealing with those issues (para 18).
It became standard practice for housing authorities in their decision letters to include statistical evidence of conditions suffered by street homeless persons, such as mental health issues and drug/alcohol problems, to argue that the presence of these factors did not differentiate an applicant from the ordinary homeless person. The test of vulnerability now required a comparison between a subset of an already vulnerable population, the homeless, to see who among them was the most vulnerable.
Vulnerability – a reboot
Perhaps rather surprisingly in hindsight, the highest court in the land did not examine the issue of vulnerability until the appeals in Hotak v Southwark LBC and other appeals  UKSC 30
;  AC 811; July/August 2015 Legal Action
50 nearly 40 years after the H(HP)A 1977.
The Supreme Court held that the relevant comparator was the ordinary person if in need of accommodation/homeless, and not the ordinary person who was actually homeless. A person was accordingly vulnerable if they were ‘“significantly more vulnerable than ordinarily vulnerable” as a result of being rendered homeless’ (para 53). Further, the assessment of vulnerability was a contextual and practical assessment and in carrying out this assessment the authority must disregard their resources and their burden of homeless people.
Lady Hale pointed out the dangers of ‘[g]lossing the plain words of statutory provisions’, stating that the position had now (wrongly) been reached where ‘decision-makers were saying, of people who clearly had serious mental or physical disabilities, that “you are not vulnerable, because you are no more vulnerable than the usual run of street homeless people in our locality”’ (para 91).
The Hotak decision itself raised questions that led to a number of further appeals. In particular, the Court of Appeal has been asked to address:
•Has the Pereira test been disapproved?
•What is meant by significantly more vulnerable than ordinarily vulnerable?
•What are the characteristics of the ordinary homeless person?
Hemley v Croydon LBC  HLR 1 confirmed that the Pereira test had been disapproved in Hotak. Importantly, the Court of Appeal held that the legal test had been changed in England so that it was ‘easier for a person to satisfy the test of vulnerability’ (para 10).
In Panayiotou v Waltham Forest LBC  EWCA Civ 1624
;  HLR 48; December 2017/January 2018 Legal Action
42, Lewison LJ held that the use of the word ‘significantly’ had not introduced a quantitative threshold to be crossed; rather, it was being used ‘in a qualitative sense’, ie, the question was would the applicant ‘suffer or be at risk of suffering harm or detriment which the ordinary person would not suffer or be at risk of suffering such that the harm or detriment would make a noticeable difference
to his ability to deal with the consequences of homelessness’ (para 64, emphasis added).
In Rother DC v Freeman-Roach  EWCA Civ 368
;  HLR 22; April 2018 Legal Action
39, it was held that the review decision could not be faulted if it failed to list the attributes of the ordinary person if homeless. However, subsequently, Guiste v Lambeth LBC  EWCA Civ 17582See also page 29 of this issue.
made it clear that the comparison is to be made with an ordinary person who is in normal health and does not have any other physical or mental illness, or disability of the type that might render them vulnerable within the meaning of the HA 1996.
Guiste also stated that it was no part of the vulnerability test for the applicant to have to satisfy the authority that their functionality would be affected, so as to make a noticeable difference to their ability to deal with the consequences of being homeless.
The treatment of medical evidence has been a recurring issue in vulnerability cases and deserves its own subheading. It was, and remains, common for housing authorities to provide medical records and any other evidence provided on behalf of a homeless applicant to a medical adviser (eg, NowMedical), who will normally carry out a summary paper-based exercise without examining the applicant or talking to their treating physicians before reaching a conclusion on vulnerability.
The dangers of such an approach were noted by Sedley LJ in Shala v Birmingham City Council  EWCA Civ 624
;  HLR 8, where he stated that although it was ‘entirely right that local authority officers, themselves without any medical expertise, should not be expected to make their own critical evaluation of applicants’ medical evidence and should have access to specialist advice about it’, nevertheless ‘care has to be taken by local authorities not to appear to be using professional medical advisers simply to provide or shore up reasons for a refusal’ (para 19).
Importantly, Sedley LJ stated that ‘[a]bsent an examination of the patient, [NowMedical’s] advice cannot itself ordinarily constitute expert evidence of the applicant’s condition’ (para 22) and that if a doctor advised on the implications of other doctors’ reports without examining the patient then that absence of examination needed to be taken into account. He suggested that authorities might need to consider whether to ask the applicant to consent to their having their own examination, or whether the local medical adviser, with the patient’s consent, might speak to the applicant’s medical adviser about matters that need discussion.
Recently, Guiste v Lambeth LBC has restated the importance of medical enquiry in vulnerability cases, and the need for proper and fair assessment of the medical evidence obtained.
Despite Sedley LJ’s cautionary comments, the paper-based assessment of vulnerability by medical advisers has continued. Recently, Guiste v Lambeth LBC has restated the importance of medical enquiry in vulnerability cases, and the need for proper and fair assessment of the medical evidence obtained. In this case, there was evidence provided to the authority that Mr Guiste had a history of depressive illness leading to acts of self-harm and/or attempted suicide at times of high stress, or at least to the contemplation of such acts, and that there was a risk that homelessness might significantly increase the probability of Mr Guiste carrying out such acts, with or without fatal consequences.
It was noted by the Court of Appeal as being relevant that:
•Mr Guiste had not been examined by any qualified psychiatrist instructed on behalf of Lambeth, which had outsourced the provision of medical advice in housing cases to NowMedical;
•although two psychiatric advisers employed by NowMedical had prepared reports for Lambeth, neither of them had interviewed Mr Guiste or examined him in person;
•in the absence of an examination of the patient, the advice from the two psychiatric advisers could not itself ordinarily constitute expert evidence of the applicant’s condition (following and applying Shala);
•the two psychiatric advisers’ qualifications were not at the same level as that of the psychiatrist relied on by Mr Guiste, Dr Freedman. Each of them was a member, not a fellow, of the Royal College of Psychiatrists; and
•the two NowMedical psychiatrists who considered Mr Guiste’s application on behalf of Lambeth did so without examining him, and without (for whatever reason) taking up the invitation to discuss his case with Dr Freedman.
Accordingly, if Dr Freedman’s evidence was going to be rejected, there had to be a rational explanation for so doing. This the review decision, even if analysed benevolently, had failed to do.
In a time of severely restricted social housing and serious budgetary constraints, the issue of whether an applicant is vulnerable will continue to be a battleground. However, recent decisions give cause for guarded optimism that the courts are willing to examine decision-making without being unduly reverential. The recent decision in Guiste may cause authorities to revisit the use of medical advisers, which, in this writer’s opinion, is a reconsideration long overdue.