Having worked for over six years in the social care sector, working with clients who have misused substances; the researcher encountered much diversity, both with the people he worked with, and the issues they experienced ranged.
In March 2007, an article was published which claimed that most people who take drugs are not likely to cause harm to themselves or others, stating that Britain's drug laws are driven by moral panic (Metro News, 9th March, 2007). Letters followed that week in which a Social Worker made claims that they have used illegal drugs such as crack cocaine and never experienced any psychiatric symptoms or ever been diagnosed with a mental illness.
Following from this, other newspaper articles relating to this area provoked interest. One particular article was based on a significant increase in the number of hospital admissions of people with a mental illness triggered by their misuse of drugs. The article examines this increase and discusses possible links between the existence of a stronger form of marijuana, known as ‘Skunk' and an increase in drug induced psychosis (Metro News, Tuesday 24th April, 2007). Therefore with professional experience and a genuine interest in the subject of dual diagnosis along with the secondary information contradicting evidence on the subject, the researcher decided that his hypothesis for research would be: People who misuse substances are more likely to experience or be diagnosed with a mental illness.
With the hypothesis defined, the researcher decided on his approach. Recognising that he was a first time researcher, he decided that he would explore the subject using information he already knew from his work within a detoxification unit, along with the secondary research he had already obtained. He also identified that to establish a new line of investigation he would also need to obtain primary evidence.
Further discussions pertaining to this topic are continued in the following slides:
To find out which primary research methods would prove most useful, the researcher decided to access the Internet. He found a website which provided research for organisations working with children and families. It was noticed that research for Lancashire Social Services on children's rights services used an anonymous questionnaire to obtain their results, whereas Leicestershire Social Care Services used a semi-structured interview to obtain evidence for the usefulness of their service (http://www.rip.org.uk/).
Following this, the researcher considered that if he was to use a questionnaire he would be able to offer anonymity to the respondents, which was important due to the sensitive nature of the research being conducted. He then determined that he would use this method with a selection of open and closed questions dividing them equally into two sample groups, determining a quota sampling method. This method would involve ensuring that half the respondents were male and half were female. Thus providing the researcher with a good comparison of the characteristics of the groups, as well as alleviating any potential reliability or validity concerns and taking into account any variables that an individual's gender may have.
It was then determined that one sample group (B), would consist of ten individuals (five male and five female) that accessed the detoxification service in which the researcher worked. The other sample group (A) would be a random group of ten individuals (five men and five women), who worked at a call centre with no identified mental illness or substance misuse history. The reason for this division of the sample groups was that it might provide a better demographic along with a more reliable and valid result, for example if all respondents were from group B and only twenty percent had a mental illness, is that because of substance misuse or a coincidence?
Alongside this, the researcher considered other methods to provide more primary data as well as alleviating the disadvantages of using the questionnaire. To achieve this he examined the research methods used by Malinowski in his covert observation of the Trobriander tribe (http://www.wikipedia.org.uk/). From this, the researcher considered the possibility of a covert observation or a case study with sample group B. However, this would breach confidentiality and be unethical, as trust would be abused and rights to decide to take part or not would be infringed. Therefore, overt observation was considered, again with sample group B. The researcher then realised that the very nature of the research could have a negative effect on some service users.
The researcher finally decided that in conjunction with his questionnaires, he would conduct an unstructured interview with a mental health nurse that worked at the detoxification unit. He considered the advantages of how an unstructured format would enable flexibility and not restrict time. However, the potential of the interview becoming an unproductive conversation could prove to be a disadvantage. Also, reliability may be brought into question, as the interviewee was a work colleague who may provide biased responses. Therefore with consideration to these factors, the researcher decided to remain with an unstructured interview, but ensure that all subject matter would remain relevant and a recording of the interview would be provided as evidence, ensuring reliability.
On formulating the questionnaire, the researcher defined his objective and took into account the style of questions he would use, deciding on a mixture of open and closed questions to enable both qualitative and quantitative responses. He ensured that all questions would be relevant to the hypothesis and not lead the respondent, thus ensuring reliability and validity.
The researcher decided he would include age and gender as two of his questions. This would enable the quota sampling method and provide information whether these variables would affect the hypothesis. Further questions determined included asking the respondents about their use of alcohol, drugs and smoking, including whether stress contributed to an increased usage or whether any negative side effects presented. The reasoning behind this was that it could indicate that a person would misuse a substance leading to an outcome that supports the hypothesis.
Other questions that were decided included whether the respondent has been diagnosed with, or feels they have had any mental illnesses. The rationale behind these was that it should support responses from other questions and clearly ascertain whether people who misuse substances are more likely to experience mental illness.
The researcher identified that group B were more likely to experience a negative effect from this research due to the nature for their need of the service. Therefore to minimise this and any confidentiality issues, it was determined that every respondent from both groups would be approached discreetly and taken to a private area in which time would be given to go through each question, with an option to leave at any time or omit any questions they felt were too personal. Finally the researcher determined that time would be given after completing the questionnaire to provide necessary support if needed.
For the interview the researcher considered both open and closed questions, but felt that if he asked any closed questions it would be too restrictive to the subject. Therefore it was decided that five open questions (taking into account confidentiality and possible bias) would be asked, providing qualitative data, including a direct question pertaining to the hypothesis.
Once the researcher had designed the questionnaire with sixteen questions, he conducted a total of three initial pilot studies amongst his peers. This was important as it demonstrated how long it took to complete the questionnaire and whether the format would effectively provide clear data and also if it was relevant and easily understood.
On analysing the results from the pilot study, it was noted that some of the language used throughout the questionnaires was not easily understood. Also a suggestion to elaborate on a few of the closed questions, allowing for a more qualitative response was made. Further suggestions included adding another variable about whether respondents believe that caffeine is an addictive drug.
From these results, the researcher accepted the suggestions made. Alterations included further questions, incorporating two more open questions and changing the language to a more appropriate level. However, one suggestion was not accepted, as the researcher believed that respondents would understand the term ‘substance misuse', given the subsequent questions. Finally, a further pilot study was conducted and final draft completed.
Following from this, the two sample groups were provided with questionnaires for completion. Then the interview was carried out and recorded following appropriate guidelines with provision of formal letters before and after the research was conducted.
With the primary research conducted, a review of the secondary information was needed. The researcher decided to do further reading on the subject and found that dual diagnosis was first recognised in America in 1993 and for the first time a possible correlation between mental illness and substance misuse was identified (Hussein Rassool, G, 2006). In addition to this, several articles in the newspapers provided further investigation. One article claimed that a person who smokes just half a ‘joint' is more likely to exhibit psychiatric symptoms similar to schizophrenia (Metro News, Tuesday 1st May, 2007). However, it was noted that no definitive link was made as to a diagnosis of mental illness, just a claim that only psychiatric ‘symptoms' would present.
One significant secondary source provided further information supporting the hypothesis by discussing the relationship between substance misuse and mental health. It was found that according to the Department of Health the use of substances would make a mental illness worse and any addiction to these substances would lead to the presentation of psychiatric symptoms. It then goes on to discuss a debate as to the extent to which a substance such as cannabis can cause a mental illness and examining the possibility of biological factors such as genetic vulnerability, thus making the risks greater for some than others (http://www.turning-point.co.uk/).
Further to this, secondary data sources showed that in the United Kingdom a third of people seeking treatment and accessing mental health services have a history of misusing substances. Also, around half of substance misuse service users have a mental health problem, which is commonly depression or a personality disorder. In addition to this it was found that male prisoners were particularly susceptible to experiencing both mental illness and addiction to a substance with seventy nine percent of drug dependent prisoners being diagnosed with two mental illnesses (http://www.turning-point.co.uk/, pg 2). However, the researcher considered that this secondary data only represented those who have sought treatment.
Therefore it appears from this information that there is a significant increased risk of mental illness if a person misuses substances. However, following confidentiality guidelines, the researcher examined the history of some of the service users, and found that many General Practitioners fail to make a link between mental illness, and an addiction, opting for a singular diagnosis, such as alcohol dependency syndrome (a form of alcohol addiction) or a mental illness such as schizophrenia. Thus treatment is usually aimed at that singular diagnosis and not both. With this apparent contradiction, the primary research would now prove invaluable.
When conducting the questionnaires all ethical considerations were taken into account and the method determined beforehand was adhered to. This preparation proved useful as some of the answers that were obtained represented very sensitive information. One particular respondent became upset and wanted to discuss his responses at great length following completion of his questionnaire. This was taken into account when planning the research methods and obviously the researcher allowed time to discuss the issues at hand with each respondent individually.
In analysing the primary data, the researcher firstly displayed all the information onto two tables, one that represented responses given by group A (fig 1.1), the other representing group B (fig 1.2). This made it easier in extracting the required information. The following two pages represent all responses received from the questionnaires.
Group A

Fig 1.1 This represents both qualitative and quantitative responses received from group A. The blank boxes represent either a non applicable or negative response.
Group B

Fig 1.2 This represents both qualitative and quantitative responses received from group B. The blank boxes represent either a non applicable or negative response.
When looking at both fig 1.1 and fig 1.2 it is apparent that most mental health diagnoses (Q16) occur within group B. Also it is interesting to see that in the responses to the mental health problems experienced (Q17), sixty percent of respondents stated depression or a personality disorder. Also, unwanted side affect (Q10) from group B tends to correlate to the mental illness experienced. These responses support the secondary information on which mental illnesses are more likely to occur in individuals who misuse substances as well as the newspaper article on psychiatric ‘symptoms' as opposed to mental illness (Metro News, Tuesday 1st May, 2007).
To analyse the quantitative data and see if there is a potential correlation between the two groups or between substance misuse and mental illness, it is necessary to breakdown the information further. This is seen in fig 1.3:
(Fig 1.3)

Fig 1.3 This represents the quantitative responses given by both groups with negative responses omitted.
From the information provided in fig 1.3 it is possible to see that only one person from group A has ever been diagnosed with a mental illness (Q16) and that was after (Q18) drinking alcohol (Q12). This person had never used any other substances and as they stated they had never increased their alcohol intake when stressed (Q13) there is a possibility that there is no correlation for this person's diagnosis of depression (Q17).
There appears to be no correlation between substances used and mental health in the responses received from group A as only ten percent of respondents stated they have experienced mental illness and none have abused drugs excluding cannabis and alcohol. No further analysis of group A would be needed at this point.
Group B's responses from fig 1.3 provide a potential correlation between any substances the respondent may have used (Q4, 5, 7, 8,9,12 and Q13) against any mental health issues experienced (Q10 and Q16). However to provide a more definite and clearer picture the information needs to be broken down further.
The following tables represent the breakdown for each section of the questionnaires from sample group B:
Group B - (Fig 1.4)

From this table it appears that there is a possible correlation between smoking more when under stress (Q4) and mental illness (Q16). However, this is only apparent amongst the males from this sample group. Therefore, due to stress being considered contributory to mental illness, this data is unreliable.
Group B - (Fig 1.5)

Fig 1.5 examines the percentages of cannabis (Q6) usage of group B. This shows that a larger percentage of males aged twenty-six to thirty-four and females aged fifty plus smoked cannabis socially. Whereas males, aged thirty-five to fifty and females, aged eighteen to twenty five, smoked cannabis heavily.
Group B - (Fig 1.6)

Fig 1.6 represents the information from differing questions and compares them against any mental illnesses (Q16). Question 7 pertains to Drug use, excluding cannabis and Question 14 asks if that person has had an alcohol addiction. On examining fig 1.6 it is apparent that 100% of all female respondents have been addicted to alcohol and been diagnosed with mental illness.
In addition, fig 1.6 clearly shows a correlation between alcohol addiction and mental illness along with a potential correlation of other drug usage. On examining the results between fig 1.2 and 1.6 there also appears to be a link between cannabis and mental health, which would support the secondary information, provided by Turning Point and original newspaper articles.
Questions eleven and sixteen represent unwanted side effects and any mental illnesses experienced. In analysing this qualitative data it is interesting to see that the types of mental illnesses correlate to that found in the secondary research. Also from the answers given it is clear that this research supports the information provided both in the newspaper articles and from Turning Point.
However, on analysis, only thirty percent of respondents put stress or coping as a reason as to why they began misusing substances. This conflicts to a degree with secondary information as this states that people with mental illness will more often seek substances to alleviate their symptoms (www.turning-point.co.uk).
One significant comparison was that between the two sample groups, group B did present with a majority of respondents being diagnosed with a mental illness and a majority of those were female. This could demonstrate that women are more likely to misuse substances and experience mental illness than men. However out of both sample groups, there appears to be a relationship between substance misuse and unwanted psychological side effects (Q10) amongst men as opposed to women. This is seen in fig 1.7:
Both Groups - (Fig 1.7)

In comparison of both groups, the questionnaires objectives have met the initial aims of the research by enabling the researcher to begin to identify that there is a connection between misusing substances and mental health, thus supporting secondary research. However, a possible bias from clients providing answers they feel the researcher wishes to include or, the possibility that respondents don't wish to discuss personal sensitive details has provided limitations and questions the reliability on the information obtained. In addition to this, some questions have proved to be not as useful as others. For example, questions about caffeine have had no relevance to the hypothesis, thus limiting results.
Furthermore, on analysing the results it is apparent that the questionnaire was too generalised and that only about eight questions were actually used to ascertain the hypothesis, providing further limitations to the research. However, the interview provided a less generalised response.
The interview followed the format designed and provided a wealth of qualitative data that would be useful for this research. Whilst conducting this, the interviewee wished to stop the tape recording at certain points to allow him to consider his answers. The researcher allowed this, despite the potential for reliability to be questioned, as he had failed to take into account the interviewees own discomfort of taking part.
The questions that were asked during the interview were in accordance with testing the hypothesis. In analysis of the qualitative information obtained, the researcher established that the interviewee was a mental health nurse and worked at the detoxification unit. This unit provided both health and social care for clients detoxifying from substances (alcohol, drugs or both). The interviewee did not directly state that people who misuse substances are more likely to experience mental illness, however his definition of the mental illnesses that are more likely to be present at the service did reflect that found from the questionnaires and from the secondary research. One interesting factor raised during the interview, was the interviewee's frank discussion of his own experiences from smoking cannabis. He stated that he experienced paranoia and hallucinations from smoking the substance. This supports the newspaper articles on this, referenced earlier. This information provides an overview of the interview; the tape recording for this is included.
In analysis the interview proved to be a very useful tool as the experience and knowledge the interviewee possessed provided answers in relation to the hypothesis. The information provided from this supported both the questionnaires and the secondary research; therefore, positively testing the hypothesis. However, the limitations for this were that the interviewee only worked with clients who have alcohol misuse issues therefore his experience would be limited to enable him to answer the questions in relation to drug misuse. Also, the factor of having to stop the tape can question the reliability as it could be assumed that the researcher led the interviewee in his responses.
Throughout this investigation, the researcher has encountered many challenges. Having not conducted any form of research before, this investigation has developed his skills in communication, both written and orally, allowing for self-confidence in his abilities to expand as the project progressed. This was identified in the interview as the researcher was very nervous, but due to the development of skills from trying out new methods and ideas as part of the planning, he was able to treat the interviewee with respect and appropriately obtain the required outcomes. Also, extracting statistics from sources such as the questionnaire has developed analytical skills and an ability to assess the reliability of information.
In conclusion, the research conducted mostly supported the secondary research. Therefore, there does appear to be a correlation that supports the hypothesis. This is evidential from both the quantitative data obtained from the questionnaire and the qualitative information provided from the interview. From this the researcher is able to ascertain, that people who misuse substances are more likely to be diagnosed with or experience mental health issues. However, there are significant limitations to the extent that this can be definitive proof of the hypothesis.
The secondary information provided from the department of health only represented the statistics of people who had sought treatment or who had been in prison. Although, the mental illnesses that appear characteristic to substance misuse did correlate to the results obtained from the primary research. This could suggest that the hypothesis is correct. Therefore this research could be used to provide a better understanding of the reasons leading to people misusing substances, as well as providing more public recognition and support services aimed at service users with a dual diagnosis.
The main issue that the researcher encountered throughout the investigation involved the scale of research that would be needed to test the hypothesis. Once established, it was quickly found that researching the hypothesis using the subject of substance misuse as a whole, in the time frame available, would be unachievable. Therefore if the researcher were to undertake a similar investigation in the future, he would need to examine the subject at a more manageable level. This would be achieved by investigating mental illness with one specific substance, such as alcohol, thus interviews and questionnaires could be directed at just one specific area.
In addition to this, future research could examine ethnicity in relation to a specific substance misuse issue. The results from this could then be used to ascertain if there is any possible genetic or cultural correlation with substance misuse. This is not an area the researcher identified in his secondary sources; therefore it would prove an interesting project to undertake in future, using his newly acquired skills from this investigation.
- Hussein-Rassool, G, (2002), Dual Diagnosis, 1st Edition, Blackwell Science, Oxford
- 9th March 2007, Manchester
- 24th April 2007, Manchester
- 1st May 2007, Manchester
- http://www.rip.org.uk/changeprojects/niftymembership.asp, May 2007
- http://en.wikipedia.org/wiki/Malinowski, May 2007
- http://www.turning-point.co.uk/NR/rdonlyres/D7775B51-37B0-4CD4-BCED-EE0177F2C454/0/Dual diagnosis_toolkit.pdf, May 2007





