I think the paper is highly cited because it introduced a novel way to examine data on the lifetime prevalence of mental disorders and presented new data which is based on a recently completed national survey. The novel feature of the analysis was that it distinguished lifetime prevalence-to-date from lifetime projected risk.

Prevalence-to-date refers to the proportion of the population who have ever had a particular mental disorder as of the point in their life when they were assessed, whereas lifetime risk refers to the proportion of the population who either already had or will at some time in the future have the disorder.

Photo: copyright 2004 Joshua Touster Photo: copyright 2006 Joshua Touster “The results of this paper raise the question of whether earlier intervention with mild cases in childhood or adolescence might be effective in preventing the progression to serious mental disorders in adulthood..”

The estimation of lifetime risk requires us to make projections into the future based on information obtained from analyzing age-of-onset distributions. We discovered that estimated lifetime risk of most mental disorders is only slightly higher than prevalence-to-date. This finding is accounted for by the fact that the age-of-onset distributions showed most mental disorders to have onsets fairly early in life.

For example, of all people who will ever have an anxiety disorder, half experience first onsets by age 11. We also showed that the vast majority of first onsets of mental disorders later in life occur in people who already had a history of some other mental disorder.

This means that it is largely comorbidity—i.e., the occurrence of multiple disorders—rather than onset of a first mental disorder that continues to grow in prevalence throughout adulthood. This finding is important because comorbidity is known to be associated with severity and persistence of mental disorders, highlighting the public health importance of interventions aimed at the prevention of the onset of temporally secondary conditions among people who already have a mental disorder.

As noted above, the paper used a novel methodology that compared lifetime prevalence-to-date with lifetime risk. The methodology itself was not new, but it was used in a novel way in order to think through the public health implications of cumulating lifetime risk of multiple mental disorders across the life span.

As with so many areas of epidemiological investigation that capture the imagination of clinicians and policy researchers, the innovation here was to look at relatively familiar patterns in new ways.

We have known for many years that mental disorders often have early ages of onset and that comorbidity among mental disorders is common, but no one had ever carried out a systematic analysis of the implications of these observations for the developmental life-course epidemiology of mental disorders.

The paper is significant because it shows clearly that mental disorders are fundamentally disorders of youth—in the sense that first onsets usually occur in childhood or adolescence. First onsets in adulthood typically are temporally secondary disorders that occur in people who have a history of prior mental disorders.

The main implication of these findings is that interventions aimed at prevention or early treatment need to focus on relatively mild cases among youth, a focus that is quite different from the current focus of most mental disorder intervention on serious cases among adults. If taken seriously, the results of the paper could lead to a fundamental reorientation of the design and implementation of interventions for mental disorders.