Survey (including pre-test probabilities)
- Submitted by:Jenifer Barrie - Student
- Institution:University of Manchester
- Date submitted:30th November 1999
1
Objectives and hypotheses
1.1
Are the objectives of the study clearly stated?
Vaguely mentioned in the abstract. 3 major aims listed in the introduction.
1.To estimate prevalence and frequency of non-fatal overdose amoungst heroin users.
2. To ascertain the circumstance in which heroin overdoses occur.
3. To examine beliefs of users about overdose.
1.To estimate prevalence and frequency of non-fatal overdose amoungst heroin users.
2. To ascertain the circumstance in which heroin overdoses occur.
3. To examine beliefs of users about overdose.
2
Design
2.1
Is the study design suitable for the objectives
Yes, survey required to obtain the users views on overdose. Data about overdose relied on self report.
2.2
Who / what was studied?
329 heroin users recruited from all areas of Sydney. Half of the subjects were engaged in treatment for opiod dependence.
2.3
Was this the right sample to answer the objectives?
Yes.
2.4
Did the subject represent the full spectrum of the population of interest?
Yes, as the sample was divided into users in treatment and those not in treatment to obtain users at different stages.
However only injectors were studied. The sample was comparable, in terms of key demographic variables, to other samples of users in Australia and elsewhere.
However only injectors were studied. The sample was comparable, in terms of key demographic variables, to other samples of users in Australia and elsewhere.
2.5
Is the study large enough to achieve its objectives? Have sample size estimates been performed?
Yes the sample size was large but sample size estimates had not been performed.
2.6
Were all subjects accounted for?
Yes, numbers do not add up in table two but this is explained by the 14% who couldn't recall the circumstances.
2.7
Were all appropriate outcomes considered?
yes
2.8
Has ethical approval been obtained if appropriate?
No mention outright of ethical approval but subjects gave informed consent and the interview took place at a venue of their choosing.
2.9
What measures were made to contact non-responders?
All patients contacted and visited personally so no non-responders.
2.10
What was the response rate?
100% Although there were missing results due to subjects bieng unable to remember some circumstances.
3
Measurement and observation
3.1
Is it clear what was measured, how it was measured and what the outcomes were?
Yes, as shown in tables 1, 2 and 3. Prevalence of overdose was measured along with the circumstances and concomitant use of other substances.
3.2
Are the measurements valid?
yes.
3.3
Are the measurements reliable?
Measurements relied on subjects accurate and honest recall. For last overdose a set definition of overdose was given for clarity.
3.4
Are the measurements reproducible?
yes, a similar survey could be sent out to a wider sample.
4
Presentation of results
4.1
Are the basic data adequately described?
Drug use history is described as mean no. of years with S.D and range stated. Polydrug use was descibed with a median no. of subjects who had used other drus and the types used.Overdose prevalence was summrized in percentages, as were the circumstances surrounding overdose.
4.2
Are the results presented clearly, objectively and in sufficient detail to enable readers to make their own judgement?
Yes in four tables as basic numerical values and percentages, with each variable clearly stated.
4.3
Are the results internally consistent, i.e. do the numbers add up properly?
Not in table two. That was explained by the 14% who could not recall circumstances surrounding last overdose. The rest of the results were consistant and numbers added up.
5
Analysis
5.1
Are the data suitable for analysis?
yes.
5.2
Are the methods appropriate to the data?
yes. For continuous variables t-tests were performed.
Categorical variables analysed using chi squared test to calculate corresponding odds ratios and confidence intervals.
Categorical variables analysed using chi squared test to calculate corresponding odds ratios and confidence intervals.
5.3
Are any statistics correctly performed and interpreted?
Yes odds ratios were given for variables more likely to lead to heroin overdose. The use of alcohol was emphasized but the OR was only 1.01.
6
Discussion
6.1
Are the results discussed in relation to existing knowledge on the subject and study objectives?
yes, circumstances surrounding overdose found in this study are compared to another study with similar findings.
Some results are not compared to other studies though, and the discussion does not always refer back to the original objectives.
Some results are not compared to other studies though, and the discussion does not always refer back to the original objectives.
6.2
Is the discussion biased?
There is no mention of possible flaws in the study and how these were overcome. When comparisons are made to other studies similar results have been found, this may, however, be because results with confounding studies do not exist or have not been looked into.
6.3
Can the results be generalised?
The use of subjects at different stages of heroin use has certainly made the sample wider. It could only be generalised to injectors, but not all users only inject.Sampling was not set so that only those who had overdosed were interviewed, an important point when generalising.
7
Interpretation
7.1
Are the authors' conclusions justified by the data?
yes, a high prevalence of non fatal overdose was found, around 2/3 of the sample. Overdose isn't related to new heroin use as found by the median time since last overdose which was 30 months.It states that alcohol is a signifant predictor of overdose but the odds ratio is small. Other conclusions are drawn according to data and are accurate.
7.2
What level of evidence has this paper presented? (using CEBM levels)
Level two
7.3
Does this paper help me answer my problem?
It considers several factors that result in overdose , but more so the frequency of, and circumstances surrounding overdose.
8
Implementation
8.1
Can any necessary change be implemented in practice?
Implementation of education for heroin users about how to prevent overdose and avoiding concomitant use of benzodiazepines and alcohol.
Emphasis to users about not injecting when alone and changes in tolerance after release from prison could also lead to harm reduction.
Emphasis to users about not injecting when alone and changes in tolerance after release from prison could also lead to harm reduction.
8.2
What aids to implementation exist?
Community drug teams and staff involved in methadone maintenance treatment, as well as GPs in giving advice to users,
Prison services and parole officers could also be aware of the risk of overdose after release from prison and educate prisoners.
Prison services and parole officers could also be aware of the risk of overdose after release from prison and educate prisoners.
8.3
What barriers to implementation exist?
Addicts who are not part of a treatment programme and would not actively seek medical service would not receive this advice.