Victimisation and its relationship to drug misuse, common mental disorder and well-being in England and Wales: year ending March 2021

This publication reports on the prevalence of drug misuse, symptoms of depression and anxiety, and personal well-being in victims of crime and non-victims across England and Wales for the year ending March 2021. Data from the Telephone-operated Crime Survey for England and Wales showed that:

• approximately 1 in 15 (6.5%) adults aged 18 to 59 years and 1 in 7 (15.0%) adults aged 18 to 24 years had taken a drug in the previous year

• cannabis was the most commonly taken drug in the last year (reported by 1 in 17 or 5.8% of adults aged 18 to 59 years) followed by cocaine (reported by 1 in 48 or 2.1% of adults aged 18 to 59 years)

• symptoms of depression in the two weeks prior to being interviewed were reported by 1 in 8 respondents (13.3%) while 1 in 5 (18.9%) reported symptoms of anxiety

Analysis of the relationships between victimisation, common mental disorder and personal well-being indicated:

• being a victim of any crime (excluding fraud and computer misuse) was associated with an increased risk of reporting symptoms of anxiety and depression

• being a victim of fraud and computer misuse was also associated with an increased risk of reporting symptoms of anxiety

• being a victim of crime was associated with lower levels of personal well-being; life satisfaction, feeling that the things you do are worthwhile, and happiness

Any drug use

Findings from the Telephone-operated Crime Survey for England and Wales (TCSEW), year ending March 2021, showed that 6.5% of adults aged 18 to 59 years reported having taken a drug in the last year (approximately 2.1 million adults) (Appendix tables 1 and 3). The proportion of adults aged 18 to 24 years who reported taking a drug in the last year was more than two times greater at 15% (approximately 693,000 adults) (Appendix tables 2 and 4).

Findings in this publication are based on data from the TCSEW, which was set up in May 2020 to continue measuring crime while face-to-face interviewing with the Crime Survey for England and Wales (CSEW) was suspended because of the coronavirus (COVID-19) pandemic. Findings on drug misuse in this publication are not directly comparable with previous findings because of differences between these two surveys (see Section 7: Measuring the data for more information). 

Estimates from the CSEW provide a better indicator of drug misuse and the trends over time. The latest estimates are available from: Drug misuse in England and Wales: year ending March 2020.

Any drug use by personal, household and lifestyle characteristics

For the year ending March 2021, the prevalence of any drug use in the previous year was highest amongst younger age groups with 15.0% of 18- to 24-year-olds and 7.8% of 25- to- 34-year-olds reporting use (Appendix table 5).

Drug use in the previous year for 18- to 59-year-olds was also twice as high among men (8.9%) than women (4.1%).

For the year ending March 2021, the TCSEW found that the prevalence of any drug use in the last year also varied by a range of other personal and household characteristics:

• unemployed adults (12.2%) reported greater use of a drug in the last year than adults who were employed (5.8%)
• those who drank alcohol three or more days a week in the last month were more likely to have reported using a drug in the last year (12.3%) than those who drank one to two days a week (8.1%), less than one day a week (6.3%) and less than once a month (including non-drinkers, 2.6%)
• private sector renters were almost twice as likely to have reported taking a drug in the last year than owner-occupiers (10.4% compared with 5.7%)
• those living in areas classified as "Cosmopolitans" were more likely to have reported using a drug in the last year (16.0%) compared with most other area types, for example "Multicultural metropolitans" (6.6%) and "Suburbanites" (4.9%)¹

The differences between these estimates, however, may not be independently related to drug use. For example, the relationship between housing tenure and drug use is likely to be affected by age, as a greater proportion of younger people are private renters.

Use of individual drug types

Since the year ending December 1995, cannabis has consistently been the most used drug in England and Wales². Findings from the TCSEW year ending March 2021 showed that 5.8% of adults aged 18 to 59 years (approximately 1.8 million adults) reported using cannabis in the last year, a substantially greater proportion of individuals than the next most prevalent drug, cocaine at 2.1% (approximately 674,000 adults; Appendix tables 1 and 3).

Cannabis was also the most common drug used by young adults, 13.7% of those aged 18 to 24 years (approximately 633,000 adults) had reported using the drug in the last year. This was followed by cocaine, with 5.3% of young adults reporting use in the last year (approximately 246,000 adults; Appendix tables 2 and 4).

Other drug use reported by the TCSEW included ecstasy, hallucinogens, opiates and amphetamines. Ecstasy was the third most commonly used drug in the last year among adults aged 18 to 59 years (0.8%; approximately 242,000 adults), whereas hallucinogens were the third most commonly used drug in the last year among adults aged 18 to 24 years (2.7%; approximately 124,000 adults).

Notes for: Drug use in the last year

1. Supergroups of the 2011 Output Area Classification Pen Portraits for the 2011 Area Classification for Output Areas (PDF,325KB)

2. Estimates of reported drug use by drug type for the year ending December 1995 to year ending March 2020 can be found in Drug misuse in England and Wales: year ending March 2020.

Common mental disorders, such as anxiety and depression, are also measured in the Telephone-operated Crime Survey for England and Wales (TCSEW). The two-item Patient Health Questionnaire (PHQ-2) and the two-item Generalised Anxiety Disorder (GAD-2) screener were used to screen respondents for symptoms of depression and anxiety respectively, in the last two weeks.

For the year ending March 2021, almost 1 in 5 adults aged 18 to 59 years reported symptoms of anxiety (18.9%) and just over 1 in 8 reported symptoms of depression (13.3%; Appendix table 8).

The TCSEW also includes personal well-being measures that ask respondents to evaluate, on a scale of 0 to 10, how satisfied they are with their life nowadays, whether they feel they have meaning and purpose in their life, and about their emotions yesterday (for example, happiness). For the year ending March 2021, the mean score for satisfaction in life and happiness was 7.5 and for feeling that things done in life are worthwhile was 7.7 (Appendix table 7). These are similar to mean scores reported from the Annual Population Survey, Personal well-being in the UK, quarterly: April 2011 to June 2021.

The Telephone-operated Crime Survey for England and Wales (TCSEW) estimated that for the year ending March 2021, approximately 12.4% of adults aged 18 years and over were victims of any crime excluding fraud and computer misuse, 10.9% were a victim of fraud and computer misuse and 1.8% were a victim of violent crime (Appendix tables, year ending March 2021).

To understand the relationship between victimisation in the last year and self-reported drug misuse, common mental disorder and personal well-being, differences in estimates for victims and non-victims were tested to see if they were statistically significant¹. Where differences in estimates were found to be significant, regression analyses were performed to measure the strength of associations in unadjusted and adjusted models. Two adjusted models were constructed to account for the following covariates:

• adjusted for age and sex

• adjusted for age, sex, disability, household structure, employment status, financial difficulty and tenure

These covariates were selected based on prior knowledge of factors associated with poor mental health outcomes² and their availability in the TCSEW³.

Odds ratios and coefficients presented in this section show the strength of an association between being a victim of crime with drug misuse, symptoms of depression and anxiety, and personal well-being. However, as the data are cross-sectional, we are unable to explore causality, for example, whether being a victim of crime leads to symptoms of depression or whether experiencing symptoms of depression leads to increased risk of being a victim of crime.

Victimisation in the last year and drug misuse

The prevalence of any drug use in the last year was higher among victims of violent crime (14.3%) compared with non-victims (6.3%; Appendix table 9). Being a victim of violent crime was associated with more than twice the odds of reporting any drug use in the last year. However, after adjusting for age and sex, the strength of the association was reduced and was no longer significant (Appendix table 11).

There were no differences among the proportion of people who had taken a drug in the last year between victims and non-victims of any crime (excluding fraud and computer misuse) or fraud and computer misuse.

Victimisation in the last year and common mental disorder

The prevalence of depression and anxiety symptoms was significantly higher for both victims of any crime (excluding fraud and computer misuse) and for victims of fraud and computer misuse compared with non-victims (Appendix table 9).

The proportion of adults who reported symptoms of depression, within victims of any crime excluding fraud and computer misuse (21.1%), was almost twice as high as non-victims (12%). In both unadjusted and adjusted models, being a victim of these crimes was associated with around twice the odds of reporting symptoms of depression compared with non-victims.

For victims of fraud and computer misuse, the prevalence of depression symptoms was higher than for non-victims (17.4% compared with 12.7%). In unadjusted models, being a victim of fraud and computer misuse was associated with increased odds of reporting symptoms of depression (Appendix table 11). Although this association remained after adjusting for age and sex, the strength of the association was reduced after controlling for other factors in the final model and was no longer significant.

The proportion of adults who reported symptoms of anxiety was also greater for victims of any crime excluding fraud and computer misuse than for non-victims (26.4% compared with 17.7%) and for victims of fraud and computer misuse (25.2% compared with 17.9% for non-victims). Being a victim of either crime type was associated with increased odds of experiencing symptoms of anxiety by around 50%, even after controlling for other factors in the fully adjusted model.

Although the prevalence of common mental disorder was similar for victims of violent crime compared with victims of any crime, there were no significant differences in symptoms of depression or anxiety between victims and non-victims of violent crime. This is likely because violent crime is relatively rare and, therefore, there is increased uncertainty around these estimates.

The relationship between victimisation and mental health is also supported by research from Victim Support, At risk, yet dismissed: The criminal victimisation of people with mental health problems (2013) (PDF, 823 KB). While it found that adults with severe mental illness were more likely to be victims of a crime, it also found that victimisation resulted in the worsening of symptoms.

Figure 3: Odds ratios for common mental disorder by victimisation, unadjusted and adjusted models

England and Wales, year ending March 2021

Notes:

1. Variables included in the fully adjusted model are: age, sex, disability status, employment status, household structure, financial difficulty, and tenure.

2. There is no statistically significant difference between victims and non-victims where the confidence interval of the odds ratio overlaps “just as likely.

Download this chart

.xlsx

Victimisation in the last year and personal well-being

The Telephone-operated Crime Survey for England and Wales (TCSEW) year ending March 2021 showed that the mean scores for all personal well-being measures were worse for victims of crime (Appendix table 10).

Satisfaction with life

Scores for satisfaction with life were 0.4 points lower for victims of any crime (excluding fraud and computer misuse) and 0.4 points lower for victims of fraud and computer misuse, compared with non-victims (using a scale of 0 to 10). After adjusting for other factors in fully adjusted models, being a victim of these crimes was still associated with lower levels of life satisfaction but the difference in scores was reduced (to 0.2 and 0.3 points, respectively).

Victims of violent crime reported scores for satisfaction with life that were 0.8 points lower than non-victims. After accounting for other factors, the association between being a victim of violent crime and lower life satisfaction remained significant but the difference in scores between victims and non-victims was reduced to 0.3 points.

Feeling that things done in life are worthwhile

The average ratings for feeling that things done in life are worthwhile were 0.3 points on average lower for both victims of any crime (excluding fraud and computer misuse) and victims of fraud and computer misuse, compared with non-victims. In fully adjusted models, the association between being a victim of these crimes and lower scores for feeling that things done in life are worthwhile remained after adjusting for other factors.

Happiness

Happiness scores were also lower for victims of both any crime (excluding fraud and computer misuse) and victims of fraud and computer misuse compared with non-victims by 0.4 points on average (using a scale of 0 to 10). After adjusting for other factors in fully adjusted models, being a victim of these crimes remained associated with lower levels of happiness but the difference in scores was reduced to 0.3 points.

Notes for: Victimisation and its relationship to drug misuse, common mental disorder and personal well-being

1. A Pearson's Chi-square test was used for categorical outcomes and a One-Way ANOVA was used for continuous outcomes. Statistical significance is measured at the 5% level, for more information on statistical significance, see Chapter 7 of the User Guide.

2. Based on findings from Adult Psychiatric Morbidity Survey (APMS): Survey of Mental Health and Wellbeing , England, 2014.

3. Ethnicity was found to be associated with poor mental health in APMS. Although this variable was available in the TCSEW, small base sizes across individual ethnic groups meant that it could not be included in regression analyses.

Any drug

Consists of any cocaine, ecstasy, hallucinogens, opiates, any amphetamines and cannabis.

Depressive symptoms

Respondents were asked the following questions from the two-item Patient Health Questionnaire (PHQ-2):

• over last 2 weeks, how often felt little interest or pleasure in doing things

• over last 2 weeks, how often felt down, depressed or hopeless

These questions had four response options ranging from 0 (Not at all) to 3 (Nearly every day). A "depression score" was then derived by summing all responses chosen, resulting in a score ranging from 0 to 6. The higher the score, the greater the severity of depressive symptoms.

A person's PHQ-2 score sits in one of two categories:

• no or mild symptoms -this refers to a depression (PHQ-2) score of between 0 and 2 (inclusive)

• moderate to severe symptoms -this refers to a depression (PHQ-2) score of between 3 and 6 (inclusive)

Anxiety symptoms

Respondents were asked the following questions from the two-item Generalised Anxiety Disorder Questionnaire (GAD-2):

• over last 2 weeks, how often been bothered by feeling nervous, anxious or on edge

• over last 2 weeks, how often been bothered by not being able to stop or control worrying

These questions had four response options ranging from 0 (Not at all) to 3 (Nearly every day). An "anxiety score" was then derived by summing all responses chosen, resulting in a score ranging from 0 to 6. The higher the score, the greater the severity of anxiety symptoms.

A person's GAD-2 score sits in one of two categories:

• no or mild symptoms -this refers to an anxiety (GAD-2) score of between 0 and 2 (inclusive)

• moderate to severe symptoms -this refers to an anxiety (GAD-2) score of between 3 and 6 (inclusive)

Violent crime

CSEW violent crime covers a range of offence types, from violence without injury or minor assaults, such as pushing and shoving that result in no physical harm or minor injury, to wounding, where the incident results in severe or less serious injury.

More information and further definitions can be found in the "offence type" section of the User guide to crime statistics for England and Wales: Measuring crime during the Coronavirus (COVID-19) pandemic.

Fraud

Fraud involves a person dishonestly and deliberately deceiving a victim for personal gain of property or money or causing loss or risk of loss to another. The majority of incidents fall under the legal definition of "Fraud by false representation" - where a person makes a representation that they know to be untrue or misleading (for example, banking and payment card frauds and dating scams). Telephone-operated Crime Survey for England and Wales (TCSEW) estimates cover a broad range of fraud offences, including attempts, involving a loss and incidents not reported to the authorities. See the "Glossary" section of Nature of fraud and computer misuse in England and Wales: year ending March 2019 for definitions of the different fraud types.

Computer misuse

Computer misuse is when fraudsters hack or use computer viruses or malware to disrupt services, obtain information illegally or extort individuals or organisations.

Financial difficulty

The Telephone-operated Crime Survey for England and Wales asks respondents "How well would you say your household is managing financially these days?"

• living comfortably

• doing alright

• just about getting by

• finding it quite difficult

• or finding it very difficult?

Those experiencing financial difficulty is defined as those who responded: "finding it quite difficult" and "finding it very difficult".

Telephone-operated Crime Survey for England and Wales (TCSEW)

Data included in this release are sourced from the Telephone-operated Crime Survey for England and Wales (TCSEW). The User guide to crime statistics for England and Wales: Measuring crime during the Coronavirus (COVID-19) pandemic provides detailed information about the crime survey.

Estimates from the TCSEW are derived from a total of 36,801 telephone interviews conducted with household residents in England and Wales aged 18 years and over between 20 May 2020 and 31 March 2021. The TCSEW was designed to operate as a panel survey, re-interviewing respondents at three-monthly interviews. Estimates of common mental disorder and personal well-being are based on data collected from a single wave with the largest sample; questions on drug misuse were only asked to the respondent once therefore no wave selection was necessary.

Drug misuse

Findings on drug misuse from the TCSEW are not directly comparable with previous findings from the Crime Survey for England and Wales (CSEW) because of:

• changes to the survey mode; TCSEW estimates are collected via telephone and respondents may be less willing to answer questions honestly compared with previous years where respondents answered these questions in complete privacy as part of the CSEW self-completion mode

• changes to the survey sample; the TCSEW sample was derived from respondents who had taken part in the CSEW in the last two years, this sample only had a few people aged under 18 years available so only those aged 18 years and over were selected for the TCSEW

• changes to the questionnaire: a more limited set of questions on drug use were included in the TCSEW because of telephone-interviewing time constraints; only questions on the use of six drug types were asked compared with 17 in the CSEW¹; any drug use is a composite variable that combines the use of individual drugs, this combination is not the same as in previous publications 

The Misuse of Drugs Act 1971 classifies controlled drugs into three categories (Classes A, B and C), according to the harm that they cause. See Section 15: Annex: Classification of drugs, for a list of drugs that respondents were asked about in the year ending March 2020 Crime Survey for England and Wales (CSEW) and their current classification under the Misuse of Drugs Act 1971.

Symptoms of anxiety and depression

The Patient Health Questionnaire (PHQ-2) and the Generalised Anxiety Disorder Questionnaire (GAD-2) only assess the degree of an individual's depressive or anxious symptoms over the previous two weeks. These measures of probable anxiety and depression have been widely validated in a variety of populations.

Estimates from the TCSEW differ from those reported by the Opinions and Lifestyle (OPN) Survey, Coronavirus and the social impacts on Great Britain: 11 December 2020, for both symptoms of anxiety (17%) and depression (19%). This is likely because of differences in geographical coverage, data collection periods, sample design and survey mode. In addition, the OPN used the 8-item version of the Patient Health Questionnaire therefore differences in the sensitivity and specificity of the 2-item and 8-item version may have also contributed to the difference in estimates.

Personal well-being measures

For an overview of the personal well-being measures, please see the Personal well-being user guidance and Harmonised principles of personal well-being.

Interpreting estimates and associations

While TCSEW estimates are based on a large sample of the population, it should be recognised that levels of drug use are relatively low and violent crime is relatively rare. While figures and comparisons published in the release are considered to be robust, estimates need to be interpreted with care and consideration.

Only differences that are statistically significant at the 5% level are described within the article, unless specifically stated that they are not significant. In the tables these changes are identified by P-values.

To understand the relationship between victimisation with drug misuse, common mental disorder and personal well-being, regression analysis has been used to measure the size and strength of the relationship between two variables, while holding all other variables in the model equal. While regression analysis can tell us the strength of the relationship between one variable and another, it cannot tell us about causality.

The two techniques used in this article were logistic regression for categorical dependent variables (drug misuse and common mental disorder) and linear regression for continuous dependent variables (personal well-being). 

The results of the logistic regression analysis are expressed as odds ratios, which is the ratio between two sets of odds (the probability of an event occurring divided by the probability of the event not occurring). The odds ratios in this article represent how personal characteristics and circumstances relate to odds of drug misuse and common mental disorder.

Linear regression results are expressed as Beta coefficients. For this analysis this is the average difference in reported personal well-being scores between those who were victims of crime and those who were not victims of crime after taking other possible influences into account.

Notes for: Measuring the data

1. The face-to-face Crime Survey for England and Wales asks the respondent if they have used: amphetamines, methamphetamine, cannabis, cocaine powder, crack cocaine, ecstasy, heroin, LSD/acid, magic mushrooms, methadone, tranquillisers, anabolic steroids, ketamine, mephedrone, other pills or powders not prescribed by a doctor or healthcare professional, other unknown smoked drugs and any other drug. The telephone-operated Crime Survey for England and Wales asks the respondent if they have used: cocaine, ecstasy, hallucinogenic, opiates, amphetamines, and cannabis.

The Crime Survey for England and Wales (CSEW) is recognised as a good measure of recreational drug use for the drug types and population it covers. However, it does not provide as good coverage of problematic drug use, as many such users may not be a part of the household resident population covered by the survey.

The Telephone-operated Crime Survey for England and Wales (TCSEW) does not cover some small groups, which are potentially important, given that they may have relatively high rates of drug use. Notably these are the homeless and those living in certain institutions, such as prisons. It also does not cover students living in halls of residence.

The CSEW includes questions about drug use in the self-completion module of the survey, which is intended to encourage honest answers, however, disclosure issues still exist around willingness to report drug use. The TCSEW did not include a self-completion module, therefore these questions were included in the main survey led by the interviewer. This would potentially exacerbate disclosure issues, where an unknown proportion of respondents may not report their behaviour honestly.

As a result, the TCSEW is likely to underestimate the level of drug misuse in England and Wales.

The CSEW provides the most consistent measures of drug use and comparisons over time. However, because of the change in mode, estimates from the CSEW and TCSEW are not comparable.

Research was conducted prior to analysis to determine what covariates should be entered into the model. Covariates found to be associated with common mental disorder, drug misuse and well-being included: age, sex, ethnicity, chronic physical condition, household structure, employment status, receipt of benefit, religious beliefs, mental health treatment.

For some variables, a proxy from the TCSEW was used: disability status (for chronic physical condition) and financial difficulty (for receipt of benefit). Ethnicity and religious beliefs are included in the TCSEW but small base sizes across individual ethnic groups and religious groups meant that they could not be included in regression analyses.

It should be noted that our regression models only partially explain the differences in levels of common mental disorder and personal well-being between people, suggesting that most of what influences these is not explained by our data. This is to be expected, as many factors that impact on common mental disorder and personal well-being are not quantified in our data sources or included in our regression models.