Letter

To the Editor: In the July issue Moser and Bond ( 1 ) raise the important question of coercion in public mental health practice. Unfortunately, their article misses several important points.

First, the word "forensic" does not appear in the article. Most states are dealing with expanding populations of justice-involved consumers in inpatient and outpatient services. For example, at the hospital where I work, the forensic population has increased from 28% in 2003 to 65% currently. Assertive community treatment teams, the article's focus, are often used as a promising approach to complex, justice-involved patients ( 2 ). It would be helpful to know the percentage of consumers in the programs studied who had court involvement, because the criminal justice system is inherently coercive. Clinical teams wishing to avoid coercive interventions often are avoiding collaboration with the justice system, exacerbating the problem of criminalization.

Second, coercion is a continuous variable, is relative, and is an important, if unstated, aspect of every relationship. Usually, the choice is not between X or not X (where X is the undesired intervention) but rather between X and Y (where both X and Y are undesired interventions).

For example, the authors erroneously state that "[d]epot injections of antipsychotic medication, often used with resistant and nonadherent consumers, virtually eliminate choice for two to four weeks." Injections may in fact be required by providers or a court as a condition of outpatient treatment. The consumer can accept the injections or refuse them and remain hospitalized.

Furthermore, even assuming for argument's sake that injection removes choice, it removes only one choice—that of taking medication. The consumer retains many other options, such as where to live and what sort of food to eat. The consumer also has the option to refuse the injection, albeit with possible adverse consequences. This is no different from an employee's refusal to comply with an undesired directive, albeit on pain of losing his or her job.

In addition, many coercive interventions require assessment of decision-making capacity (such as the ability to manage funds effectively in assignment of a representative payee) or of dangerousness (in civil commitment). The authors ignore this issue.

Third, the authors found that the strongest predictors of a team's use of coercive interventions were the percentage of consumers with schizophrenia spectrum disorders and with active substance use. Both groups of disorders are associated with poor insight and denial of illness ( 3, 4 ). Such individuals are prime candidates for involuntary interventions, and it is not surprising that these two factors predict use of coercive interventions.

Finally, the authors do not comment on outcomes of treatment provided with and without coercion for similar groups of patients. This is the most important area for future study: are coercive interventions useful in changing a person's life course? If they are, they will continue to be used, regardless of how uncomfortable the idea of coercion makes us. The authors are correct when they conclude that "excessive use of control should be cause for concern," but they leave us in the dark about the line between appropriate and excessive use of coercive interventions.

Dr. Roskes is director of forensic treatment at Springfield Hospital Center, Sykesville, Maryland.