Comment on “Tardive Dyskinesia Circa 2006”

To the Editor: Risk assessment of tardive dyskinesia might be further strengthened by considering its severity and impact on quality of life. Descriptive data from two clinical trials (1 , 2) are used to illustrate the relationship of Quality of Life Interview ratings and tardive dyskinesia levels documented by combined average scores on the three global ratings of the Abnormal Involuntary Movement Scale.

Using 1,275 ratings from 423 refractory schizophrenia patients (1) and observations when no tardive dyskinesia was present as a reference point (mean Quality of Life Interview score=43.3 [SD=17.1]), mild tardive dyskinesia was associated with a 3.6-point lower (–8%) mean Quality of Life Interview rating, and moderate tardive dyskinesia was associated with a 6.9-point lower (–16%) mean Quality of Life Interview rating. After multivariate adjustment for differences in concurrent symptoms (total Positive and Negative Syndrome Scale scores), differences in adjusted means reversed direction. Adverse effects of tardive dyskinesia on ratings of quality of life were thus not independent of the effect of symptom severity in this cohort.

Data from another trial (2) included 826 observations from 309 patients with schizophrenia. As compared with observations when no tardive dyskinesia was present (mean Quality of Life Interview score=47.2 [SD=16.7]), mild tardive dyskinesia was associated with a 4.0-point lower (–8%) average Quality of Life Interview rating, and moderate tardive dyskinesia was associated with a 2.2-point (–4.5%) lower rating. After adjusting for symptoms of schizophrenia and other neurological side effects, these differences were reduced only slightly to 3.6- and 1.8-point lower ratings (–7.4% and –3.6%, respectively).

These data show average net reduction of quality of life associated with mild or moderate tardive dyskinesia ranging from 0% (based on the risk-adjusted analyses in the first cohort) to –16% (without adjustment). John M. Kane, M.D., (3) suggests a 3%–5% annual incidence of tardive dyskinesia in patients with first generation antipsychotics and 1% with newer medications. The attributable risk of tardive dyskinesia because of first generation antipsychotics would thus be 2%–4% per year. Multiplying the attributable risk by the reductions in quality of life estimated above, one can calculate average annual tardive dyskinesia-related reductions in quality of life attributable to first generation antipsychotics. This risk would be estimated to range from 0% (based on the risk-adjusted analysis of the first cohort) to a maximum of –0.64% (i.e., 0.04 attributable risk of tardive dyskinesia multiplied by –16% reduction in Quality of Life Interview scores). These analyses are limited, however, by the absence of data on severe tardive dyskinesia.

An upper bound estimate of the average risk of decline in Quality of Life Interview scores because of mild or moderate tardive dyskinesia with first generation antipsychotics may thus be less than 1%. Clinical decision making must be based on individual patient circumstances and preferences, rather than on average risk calculations. However, side effect risks are central to antipsychotic treatment decisions, and it may be informative to use quantitative methods to evaluate their net effects on measures of well-being.