The following chapter lays out the basic data on the arrestee population studied, their self-reported drug use, and the results of the urinalyses. It also includes self-reported details of their drug use and alcohol consumption. Data from the three phases of the study were combined into one data set, as there were no major differences between phases. Figures given refer to this combined data set, but are weighted on the basis of police station and arrest category to reflect the actual flow of arrestees through the cells during the days data was collected at each police station.

Demographics

As reflected in Figures 1 and 2, the following demographic trends were noted:

As reflected in part in Figures 3 and 4, the respondents showed the following background particulars:

A total of 14% of the sample interviewed were not South African citizens and just over
one tenth of the sample was arrested for being in South Africa illegally (Figure 5). This
shows that most of the foreign nationals arrested were arrested for immigration matters
only. Of these illegal immigrants, almost three-quarters (73%) were interviewed in
Johannesburg, predominantly at the Hillbrow and Kempton Park police stations.
There are two reasons for this:

Figure 5: Nationality of the non-South African sample

Crime profile of arrestees

The offences for which the subjects were arrested were broken down into the categories of violent crime, property crime, drug-related crime, and 'other' (Table 2). 'Other' crimes constituted the single largest arrest category, with 40% of the arrestees falling under this heading.

Property crimes accounted for almost a quarter of those arrested, followed by violent crimes at 20%, with substance-related crimes accounting for only 16% of the crimes for which arrestees were held (Figures 6 and 7).

Figure 6: Categories of arrests



Figure 7: Offence categories by sites

So, who are these 'other' offenders? Illegal immigrants constituted the largest proportion of this category, accounting for almost a fifth (18%) of those interviewed in the first phase, declining to around a tenth in phases two and three (9% and 10% respectively). Aside from this group, the rest of the 'other' subjects were accused of a range of crimes, both serious and minor. Almost 5% of the respondents were arrested for fraud, and 3% for traffic offences.

This distribution does not necessarily reflect the real crime profile of each area—it is more likely to reflect the station-level priorities of the police, which can vary from week to week. For example, during one phase of the study, the Hillbrow police station in Johannesburg prioritised illegal immigrants and sex workers, which resulted in a particularly high number of both being detained over a particular weekend. Similarly, arrests for substance-related offences were lower in Johannesburg than in Durban and Cape Town, implying that police in Johannesburg are not prioritising these crimes to the same extent as the other metros (see Figure 7).

Those accused of committing violent crimes come from all generations—between one fifth and a quarter of those arrested within each age category were detained for violent crimes (Figure 8). Women arrested for violent crimes were generally accused of assault resulting in grievous bodily harm and common assault—very few were charged with murder. Those arrested for substance-related offences were older than might be expected: nearly a quarter were above the age of 36 within the male sample.

Figure 8a: Offence by gender and age (Male)



Figure 8b: Offence by gender and age (Female)

As detailed in Figure 9, white and Indian men were most likely to be arrested for substance-related crime, while African men were most likely to be arrested for violent and other crime (including illegal immigration). Women were generally more likely to be arrested in the 'other' category, which includes offences such as violations of the child care act and prostitution. Coloured females also show a high rate of substance-related arrests, which may be due to the fact that the majority were arrested in drug infested areas of the Cape.

Figure 9a: Offence by gender and ethnicity (Male)



Figure 9b: Offence by gender and thnicity (Female)

Given the links between poverty and crime and the tendency to under-report monthly income in studies of this nature, it is not surprising that the majority of those interviewed claimed to fall into the lowest income bracket (Figure 10). Over a quarter of those arrested received a monthly income of less than R200, but 17% of the sample received over R2 000 monthly. While those arrested for property crimes were more likely to fall into the lowest income category, those arrested for substance-related offences followed the broader trend.

Figure 10: Offence by monthly income level

Self-reported substance use

Respondents were asked whether they had ever tried a range of drugs, and whether they had used these within the past 12 months, 30 days, seven days, and three days. While self-reported use of tobacco and alcohol was high, reported use of harder drugs in the preceding thirty days was comparatively low (Figure 11).

Figure 11: Reported use int he last 30 days

Dagga (cannabis) is the drug that arrestees were most likely to say that they had used: 839 respondents admitted to using it in the past twelve months, of whom 738 (88%) reported using it within the past 30 days. This probably indicates both the high levels of usage in the arrestee population as well as a willingness to admit to using a drug that has some degree of popular acceptance.

After cannabis, those interviewed were most likely to admit having, at least once, used over-the-counter drugs to achieve effects other than what they are medically prescribed for. Ten percent reported such use, while 8% claimed to have done so with prescription drugs. Assuming they understood the question, this high rate of reporting could reflect an under-explored pattern of drug use, or it could be due to the fact that this behaviour is not necessarily illegal.

Mandrax (9%) and crack (3%) were the drugs the arrestees were next most likely to admit having tried at least once, and in both instances were used by almost 80% of the users in the past month. The use of amphetamines, steroids, heroin, PCP, LSD, cocaine, relaxants and designer drugs were all reported, albeit by relatively few arrestees.

Those who admitted usage were questioned further about their behaviour, including the amount of money they spend monthly on drugs. Here cocaine users distinguished themselves, with over half admitting to spending more than 40% of their income getting high. This is especially impressive given that cocaine users were far more likely to fall into the higher income brackets. Mandrax, despite its lower cost, also took a good chunk of the income of its generally poorer user base—one third said they spent more than half of their income on this addictive substance. But one third of all self-confessed drug users were more casual in their consumption patterns, estimating that they spent less than 1% of their income on drugs.

Self-reporting showed many more arrestees willing to admit alcohol than drug use. Unfortunately, given unavoidable administrative delays, it was not possible to chemically test alcohol levels. In the third phase of the study, questions were added to assess weekly drinking patterns of arrestees who reported drinking in the past 12 months. Despite a large percentage being unemployed, alcohol consumption was largely limited to weekend binges, during which nearly half drank more than five drinks a day.

Figure 12: Weekday and weekend alcohol consumption of arrestees who reported drinking in the past 12 months

Use of drugs and alcohol when committing crime

In order to probe the causal relationship between drug use and crime, arrestees were asked two sets of questions:

Were you in need of drugs (or alcohol) at the time of the alleged offence?



Were you under the influence of drugs (or alcohol) at the time of the alleged offence?

The former question was aimed at probing the relationship between substance dependency and crime, and the latter the relationship between disinhibition and crime. Less than 5% admitted to being either under the influence or in need of drugs at the time of the offence, and half of those who denied drug use at this stage later tested positive for drugs (Figure 13).

Figure 13: Percentage 'in need' or under the influence

Still, nearly 20% said they were under the influence of alcohol at the time of the alleged offence, and 16% said they were in need of a drink. This result must also be treated with some caution, however, as arrestees may have felt compelled to explain away their anti-social behaviour with intoxication. This was illustrated in the drug urinalysis results that follow, where of the 98 respondents that self-reported the use of drugs to commit crime, only 74% tested positive. While some of these may have committed the crime for which they were arrested some time ago, most police arrests occur at the time of the offence.

Urinalysis

Almost half (46%) of those arrestees that consented to interviews tested positive for one or more of the six drug types for which tests were done. The most prevalent drug found was cannabis, with almost 40% of those consenting testing positive for this drug. This was followed by Mandrax (21%), cocaine (4%), opiates (2%), benzodiazepines (2%) and amphetamines (1%). Drug usage seems to be more prevalent among young people and men, with 65% of males below the age of 20 testing positive for drugs. Women using drugs tended to be older than male users (Table 3).

Table 3a: Percentage testing positive for any drug (age)