Study finds discrepancies in prevalence estimates of mental disorders in middle-aged and older adults

Common methods of assessing mental or physical disorders may consistently underestimate the prevalence of mental disorders among middle-aged and older adults, a new study from the Johns Hopkins Bloomberg School of Public Health has found.

The analysis, led by postdoctoral fellow Dr. Yoichiro Takayanagi, and published in the January 8 online edition of JAMA Psychiatry, reveals substantial discrepancies among mid-life and late-life adults in reporting past mental health disorders, including depression, compared with physical disorders such as arthritis and hypertension.

"The takeaway is that lifetime estimates based on [participant] recall in cross-sectional surveys underestimate the occurrences of mental disorders over the lifetime," said Ramin Mojtabai, MD, PhD, MPH, MA, associate professor in the Bloomberg School's Department of Mental Health and senior author of the study.

The findings are believed to be the first to examine retrospective evaluations versus cumulative assessments among older adults. Recent studies of adolescent and young adults have also found discrepancies in prevalence estimates of common mental disorders between retrospective reports versus multiple assessments over time.

The study was based on interviews in 2004 and 2005 with 1,071 adults who had since the early 1980s participated in the Baltimore Epidemiologic Catchment Area Survey, a longitudinal study that included three earlier sets of interviews going back 24 years.

When asked to provide so-called retrospective evaluations in six categories - major depressive disorder; obsessive-compulsive disorder; panic disorder; social phobia; alcohol abuse or dependence and drug abuse or dependence - participants underreported their disorders even though they had reported them one or more times in three previous assessments. In contrast, the same cohort, when asked for retrospective evaluations of physical disorders in five categories -- diabetes, hypertension, arthritis, stroke, cancer -- provided histories that were much closer to cumulative assessments from the earlier interviews. For instance, only one out of 10 underreported that they'd previously had diabetes.

As part of the study, trained interviewers administered a structured interview that yields psychiatric diagnoses based on DSM-III or DSM-III-R criteria, in four waves of interviews. In the first two waves, in 1981 and again in the 1982, DSM-III was used. In the third follow-up, which took place in 1996, and the fourth, in 2004 and 2005, the revised DSM-III was used. Failure to recall lifetime mental disorders was defined as not meeting criteria for the lifetime history of the mental disorder in the fourth round of interviews, despite reporting symptoms that met criteria for that disorder in at least one previous interview.

The study found that the lifetime estimates of mental disorders ascertained by retrospective versus cumulative evaluations were 4.5% versus 13.1% for major depressive disorder; 0.6% versus 7.1% for obsessive-compulsive disorder, 2.5% versus 6.7% for panic disorder, 12.6% versus 25.3% for social phobia, 9.1% versus 25.9% for alcohol abuse or dependence, and 6.7% versus 17.6% for drug abuse or dependence.

In contrast, the estimates of physical disorders measured by retrospective versus cumulative evaluations were 18.2% versus 20.2% for diabetes, 48.4% versus 55.4% for hypertension, 45.8% versus 54.0% for arthritis, 5.5% versus 7.2% for stroke, and 8.4% versus 10.5% for cancer.

Dr. Mojtabai explained that the contrast between the recall of mental and physical disorders is noteworthy and may be attributable to differences in age at onset and the course of these disorders. "Stigma associated with mental disorders, as well as the fluctuating course of mental illnesses, might partly explain the discrepancies, as well as differences in ages of onset of mental and physical disorders. Mental disorders start earlier and have a higher prevalence in early to mid-life, whereas physical disorders are typically illnesses of middle and older age and tend to be chronic."

The authors noted that measurement issues might also help explain the differences in recall of mental and physical illnesses. Ascertainment of mental disorders was based on symptom criteria, while ascertainment of physical illnesses was based on the participant's report of presence versus absence of a particular physical disorder.