Watzke et al. state at the outset that the population of interest comprises German patients with severe mental illness and undergoing vocational rehabilitation. The research team may perhaps be forgiven an excess of bravado. In fact, the results of this prospective study can reasonably be projected to the universe of Germans with mental disorders only if Rehabilitation Psychisch Kranker (Institution of Rehabilitation for the Psychiatrically Ill and Disabled) in the city of Halle is a kind of secondary care, central referral institution that receives the mentally ill referred for vocational rehabilitation from all over the country. Strictly speaking, therefore, applicability to a nationwide population of interest will be true only if a second study were to replicate research findings in some other German institute.
The sample consisted of 41 patients with some mental disorder or other and, the authors declare, “symptomatically stable on neuroleptic medication, had German as their native language, were … 18 to 65 years, had no co-occurring diagnosis of substance abuse or dependence according to DSM-IV, and had premorbid IQ not lower than 70 points.” (Watzke et al., 2008: 249) One assumes that by “points”, the authors meant percentile ranking.
If so, the researchers may have been overly conservative since a meta-analysis (Woodberry, Giuliano, and Seidman, 2008) suggests that the typical mental ability deficiency on onset of frank psychosis is closer to half a standard deviation below that of normal subjects, meaning a minimum of approximately 84 points of IQ. On the other hand, the researchers were decidedly wrong to label the study concerning learning potential and rehabilitation outcomes in schizophrenia since 39 of the 41 subjects were diagnosed for some other mental disorder besides schizophrenia and schizoaffective disorders.
With respect to ethical considerations, an appraisal of the article shows that the following were satisfied:
- The researchers had the relevant training since Dr. Watzke was affiliated with the Department of Psychiatry, Psychotherapy, and Psychosomatics of Martin Luther University of Halle-Wittenberg, Dr. Kuss was with the Institute of Medical Epidemiology, Biostatistics, and Informatics of the same University, Dr. Brieger was a staff psychiatrist at Bezirkskrankenhaus in Kempten, while Drs. Schoettke and Wiedl were associated with the Department of Psychology, University of Osnabrueck. All Institutions mentioned are in Germany.
- Study participation was voluntary since two patients were allowed to withdraw prior to data-gathering.
This is a quantitative study because the sample size, though modest, exceeded the threshold required for parametric statistics, because the authors employed a standardized study instrument, and they confidently subjected the data to measures of central tendency, dispersion and the Spearman correlation coefficient.
As to type, this is a longitudinal, cross-section cohort study. Under the National Guideline Clearinghouse summaries, the design comes closest to “II-2: Evidence obtained from well-designed cohort or case-control analytic studies” although results were exclusively from one center or one patient group. No provision was made for a control group; the researchers were concerned mainly with the differential outcomes among “high scorers, learners and non-learners” or those who had high, modest and low learning potential.
Watzke et al. defined the scope of their intervention as a prospective study, thereby claiming a breakthrough since all prior research, they report, had been cross-sectional designs. The absence of a control group violates that requirement for a prospective study. However, it is also true that in epidemiology a prospective study is sometimes used as a synonym for cohort study. Nevertheless, it is a true longitudinal research because the authors followed the patients through stages of diagnoses, pretest, intervention, posttest and post-release follow-up in a work setting, the entire process lasting 15 to 18 months.
References
Watzke, S., Brieger, P., Kuss, O., Schoettke, H. & Wiedl, K. H. (2008). A longitudinal study of learning potential and rehabilitation outcome in schizophrenia. Psychiatric Services, 59 (3), 248-255.
Woodberry, K. A., Giuliano, A. J. & Seidman, L. J. (2008). Premorbid IQ in schizophrenia: A meta-analytic review. Am J Psychiatry; 165: 579-587.