|
|
South West Public Health Observatory |
|
| The impact of drug misuse on health in the South West, 1996-2001 | ||
|
|
Regional Drugs Misuse Database
The first time a person with a drug misuse problem (as defined on p.4) makes contact with an agency a notification form (K071) is completed, with the person's permission and the assurance that their identity will remain anonymous. Various agencies are requested to make notifications, such as specialised drug services, GP's, Police, Probation Service and Prisons. If the individual re-presents after six months a new notification is made.
Each form contains 55 data fields, with the person's initials, date of birth and gender, and includes information on area of residence (postcode to 4 digits and locality), ethnicity, housing, employment, substances of misuse and routes of administration. Information is also available concerning children and whether they live with their drug-misusing parents. Details are also recorded concerning the source of the notification, type of agency and who referred the person to the agency.
These fields comprise the national data set, in addition to which the South West database requests information about pregnancy, and an assessment by the clients of their age of first drug use and age of first problems because of drug use.
The completed forms are sent to the RDMD in Bristol where in order to distinguish one individual from another a match is made, based on the person's initials, date of birth and gender. A unique case number can then be assigned but confidentiality preserved. Each time an individual is notified the case is assigned an episode number. Newly notified persons may be distinguished from repeat cases by using their first episode number. Individual persons may be distinguished from each other by matching their case numbers, to avoid double counting.
The forms are often incomplete and notification practice varies considerably across the region, with some areas making few notifications. Notes concerning these inconsistencies are made in the Results and Discussion sections.
In order to achieve consistent terminology, a 'case' in this study is defined as an individual person. Where we have described 'new cases' we are referring to unique individuals, notified for the first time. Where we have described 'repeat cases' we are referring to unique individuals who have been renotified. Examining the differences between new and repeat cases is useful in looking at the differences in severity of drug misuse over time.
For this study five years data were provided by the RDMD, in Excel format and transferred for analysis to SPSS and Splus. Data years ran from April 1st 1996 to March 31st 2001.
There was limited information available regarding area of residence, partly because postcode and locality information were so incomplete, or inconsistently recorded. In order to allocate a DAT area of residence for each episode a code was assigned based on the available information from postcode or locality. It was possible to assign 89.2% of notifications in this way. When a notification had neither information on postcode nor locality, but the person had other episodes, a DAT area was assigned according to the DAT area of the closest episode in time. A further 4.9% of cases were assigned in this way. For the remaining 5.9% of cases where no such information existed, the DAT area was assigned according to the DAT area of the referring agency. In order to undertake this task we are indebted to the work of Helen Cooke and Paul Brown of the South West Public Health Observatory.
The importance of allocation of area of residence relates to assessing migration of problem drug users and cross border flow. It is quite common for residents from one area to be seen by agencies in another. It is also important to be able to quantify the numbers of people coming into the region from other parts of the country.
When examining the number of persons from each DAT area, the residency criterion was where the individual had been living at the time of notification during any data year. Therefore individuals could be counted more than once if they moved area and were re-notified by an agency in a different area. But they could not be double counted within one DAT area. So for Table 8, the numbers for each DAT area are for persons appearing in that DAT area, any time during 1996–2001, but once only. From Table 8 it is true to say that according to the RDMD there were 9,257 Bristol DAT residents notified during the five years, out of the 13,500 individual persons notified by Bristol DAT agencies. Therefore 4,711 people living outside the Bristol DAT area were seen by agencies within the area.
HES
Similar methods were applied to the HES database but numbers of admissions is shown in the text. Admissions are distinguished from episodes (whereby a case may be transferred from one consultant to another during the same admission) by using 'episode=1'. In order to distinguish individual patients from admissions each case has a unique patient identifier number. It was decided to present admission data because it gives a more accurate picture of workload, and upon examining the data it became apparent that there were few repeat cases.
Cases were selected according to ICD10 diagnostic codes. Those for narcotics self-poisoning are T40, further subdivided into Cause code X42 (accidental) and X62 (deliberate). These are the codes assigned after individuals are admitted to a hospital bed following an overdose.
Using codes F11-F19, 'mental and behavioural disorders due to psychoactive substance use', identified psychiatric cases.
Conditions and data sources
Sources of data pertaining to various health issues related to drug misuse, summarised across the life span, are presented in Table A, together with the type and level of information described in the report.