E/CN.4/2006/7/Add.2 page 16 59. The Working Group commends Canada for these innovative programmes, aimed at reducing the levels of pretrial detention. The Working Group considers that such programmes deserve being “exported” from Ontario to other jurisdictions in Canada and might prove useful as models also to other countries. At the same time, however, the Working Group remains concerned about the continuous increase of recourse to pretrial detention in Canada over the last 10 years, as explained below. E. Detention of refugee claimants and foreigners without status is the exception 60. Although the increased concern about security has had an impact in Canada, the detention of refugee claimants and foreigners upon arrival in Canada or in view of removal from Canada remains the exception. Moreover, in most cases, immigration custody lasts less than 48 hours or only a few days. IV. ISSUES OF CONCERN A. Detention in the framework of criminal procedure 1. Difficulty in obtaining bail for accused belonging to vulnerable and marginalized groups 61. While the convict population has been constantly decreasing since the enactment of the Sentencing Reform Act, the number of persons detained on remand has been growing considerably. Canada-wide, the daily count of persons detained on remand increased 9 per cent from 2001-2002 to 2002-2003.3 62. The incessant rise of the remand population (against a background of decreasing crime rates and decreasing sentenced prison population) is of great concern. First of all because under both Canadian and international law everyone has the right to be considered innocent until proven guilty at trial. Secondly, while in Québec and British Columbia persons detained on remand have access to the programmes that benefit those serving a sentence, this is not the case in most of the other provinces. Thirdly, as Canadian courts have recognized in giving double and even triple credit for pre-sentence custody, conditions of remand detention are generally harsher than those of persons serving a sentence. 63. Fourthly, pretrial detention disparately impacts on vulnerable social groups, such as the poor, persons living with mental health problems, Aboriginal people and racial minorities. In evaluating whether an accused person is likely to attend future court hearings in his case, and therefore should be granted bail, the Crown and the courts have traditionally used, inter alia, indicators relating to the accused’s “roots in the community”. These criteria (which of course are common to most bail systems), when applied to an accused person who is poor, living with mental health problems or a drug addiction, or otherwise marginalized, are likely to lead to denial of bail. 64. Another worrying aspect is the high number of persons living with mental health problems kept in pretrial detention instead of in a medical setting, where they could receive adequate treatment. Sometimes judicial orders that criminal defendants awaiting trial be