Depression was studied in a community sample of 962 males and 1555 females aged 55 years and over living in Kentucky in 1981. The sample was representative of the population in Kentucky in that age group and quite similar to that US population. The Center for Epidemiologic Studies Depression Scale was used as a measure of depression, and 13.7% of the males and 18.2% of the females were at or above a previously established cutpoint of 20 for adults over age 55 years. Significant relationships to depression were found in both sexes for age, education, income, housing quality, marital status, and health. For females, the age-depression relationship was not linear. By far the strongest relationship was with self-reported physical health. Significant proportions of those with self-reported kidney or bladder disease, heart trouble, lung trouble, hardening of the arteries, and stroke were above the depression cutpoint. For those conditions, physicians could expect high levels of concomitant depression in about one fourth of males and at least one third of females. These levels of depression were not found for those with high blood pressure, stomach ulcers, cancer, or diabetes. Over half of the sample reported taking prescribed medication and over half had needed a physician's care in the previous six months. Only 3.9% of the males and 3.2% of the females admitted to needing help for mental health problems. Thus, older adults with depression would probably be more likely to seek help from physicians than from services or professionals with explicit mental health labels.

Five scales were assessed as mental health measures for older persons: Affect Balance, The Center for Epidemiological Studies Depression Scale, General Well-Being, LSI-Z Life Satisfaction, and Trait-Anxiety Inventory. These scales were administered to a community sample of 279 older persons and a clinical sample of 109 older persons who were in psychiatric inpatient units. In both samples, the internal consistency reliabilities for the anxiety, depression, and well-being scales were moderately high to high, for the life satisfaction scale they were acceptable, but the reliabilities for the affect balance scale suggest some caution in its use. For validity, multivariate analyses of variance found that all scales significantly discriminated between the two samples. The well-being and then depression scales were the strongest discriminators while the life satisfaction scale had the weakest validity. Cutting points for the well-being and depression scales are suggested for estimating the proportions of older persons who would be probable at-risk for disorder that requires intervention.

An anxiety symptom scale was administered to a community sample of 713 males and 1338 females representative of persons aged 55 and older in Kentucky and quite similar to that of the US population in that age range. On the basis of a cutting point established previously, 17.1% of the males and 21.5% of the females were estimated to experience sufficient anxiety symptoms to place them at risk so as to require some form of intervention. In addition to gender and marital status differences, the study found a systematic but complex relationship between anxiety and age within this age range. Anxiety was inversely associated with socioeconomic status, education, and urban dwelling. Anxiety was inversely and highly correlated with physical health, and related to the presence of nine specific medical conditions and the need for and use of a number of medical services. The implications of the findings regarding the use of medical services for early diagnosis and treatment for mental health needs are discussed.

An empirical test was conducted to determine the degree to which respondents would follow the directions of the randomizing device in the directed response model of the randomized response technique (RRT). The results showed that a sizable percentage of the respondents would not give the directed response to a question if that response was a socially undesirable one and the question was sufficiently sensitive.

No abstract