Wilhelm, a psychologist affiliated with the department of Instruction technology at the University of Twente, together with three pupils of the Bonhoeffer College in Enschede, studied the effect of one can of energy drink on the outcomes of six cognitive tests. The study had a control group drinking water (n = 34), a placebo group that got sugar-free lemonade (n = 35) and the experimental group that received a caffeine drink (n = 34).

The longest cognitive test showed a statistically significant effect. This significant outcome was in the predicted direction, although the placebo group and experimental group ended up close. We are only informed that all other tests showed no differences between the three conditions. Differences in proportions and numbers of correct answers were tested using ANOVA. This statistical method, however, assumes that the dependent variable is continuously and normally-distributed with a constant variance. Proportions and numbers "by definition" do not follow a normal distribution. Possibly differences would have been found using more appropriate analyses (logistic- and poisson regression analyses). Moreover ...

Penterman and Nijman tested the predictive validity of the Checklist of Risks to the Crisis team (crc). This instrument was specifically designed to assess the risks of aggression from severely disturbed psychiatric patients before they are seen by the crisis team. Any perceived aggression following outreach contact was recorded by means of the Staff Observation Aggression Scale-Revised (soas-r).

A forward stepwise logistic regression analyses was performed. Three factors in the CRC were important: the clinical assessment on a visual-analogue scale, an estimate of the number of aggressive persons in the vicinity of the crisis patient concerned and the reporting of crises by the patient himself.

The authors conclude that a lack of association with aggression of some of the variables in the study can be informative for the work of crisis teams. However, if the outcome of the test is non-significant, this is not proof of the null-hypothesis. Pr(D|H) is not Pr(H|D) or, in other words, the absence of proof is not proof of the absence.

Penterman EJM, Nijman HLI (2009).Assessing aggression in patients of the mental health crisis service. Tijdschrift voor psychiatrie 51(2009)6, 355-364

Hamerlynck and colleagues investigated differences between two groups of girls sentenced to juvenile detention centres under civil law or under criminal law. A sample of 211 female minors was studied for differences in offence history, socio-demographic characteristics, contact with the social services, psychiatric disorders and trauma. Most girls were detained under civil law: 82% versus 18% under criminal law.

Results indicated that the ‘criminal’ group more often had a violent history of delinquency and a non-Dutch background, whereas the ‘civil’ group more often had a background of residential placements, oppositional-defiant disorder, suicidality and self-harm.

However, the researchers also interpreted non-significant results: “there were striking similarities in the girls’ behavioural problems and psychiatric disorders”. But the null-hypothesis (no differences) can only be rejected, not proven. Most researchers, however, would like to conclude something - the statistician Cohen ridiculed these nul-hypothesis tests.

Girls detained under civil and criminal law in juvenile detention centres; psychiatric disorders, trauma and psychosocial problems – S.M.J.J. Hamerlynck, L.M.C. Jansen, Th.A.H. Doreleijers, R.R.J.M. Vermeiren, P.T. Cohen-Kettenis – Tijdschrift voor psychiatrie 51(2009) 2, 87-96


Schadé and colleagues set out to discover whether post-treatment relapse can be prevented in alcohol-dependent patients who suffer from both alcohol-dependence and a comorbid anxiety disorder. A randomised controlled trial among 96 abstinent patients evaluated the effect of additional treatment specifically for the comorbid anxiety disorder. Care-as-usual consisted of an intensive psychosocial relapse-prevention programme, in the experimental condition this program was supplemented by cognitive behavioural therapy and optional pharmacotherapy.

Results showed that anxiety symptoms and alcohol relapse rate were lower in the group receiving cognitive behavioural therapy. The study design was powered to detect a drop in the rate of post-treatment relapse of 20% but the researchers found a small effect size of 4 % and a non-significant Chi-square value. Schadé et al. conclude that anxiety treatment for alcohol-dependent patients with a comorbid anxiety disorder “has no significant effect on the outcome of alcohol treatment programmes”.

Service management probably will not be inclined to invest in the programme add-on's that were tested in this study. However, the non-significant Chi-square is not proof that anxiety treatment has no significant effect on outcome: the probability of Death by Hanging differs from Hanging as cause of Death or Pr(Data|Hypothesis) <> Pr(Hypothesis|Data).

The effectiveness of anxiety treatment on alcohol­dependent patients with a comorbid disorder: a randomised controlled trial ­ A. Schadé, L.A. Marquenie, A.J.L.M. van Balkom, M.W.J. Koeter, W. v an den Brink, R. van Dyck ­ Tijdschrift voor psychiatrie 50(2008)3, 137-148

No-effect Interpretation

feed-image Follow new articles