Allan J. Schwartz

College student suicide in the United States: 1990-1991 through 2003-2004.

Suggestions that there is a growing epidemic of suicide among college students in the United States are false. The National Survey of Counseling Center Directors reports 1,404 student suicides over a 14-year period and an adjusted suicide rate of 6.5, half the rate of the general US population (12.6 for all races) during this period when matched for gender and age. Counseling centers appear effective in treating suicidal students, for although the suicide rate for students who were currently or previously clients at campus counseling centers is 3 times the rate of other students, student clients have 18 times the risk of suicide compared to students in general. Identifying and referring students at elevated risk for suicide could further reduce the crude and relative rate of student suicide. However, even programs that do this only moderately well may require substantial increases in counseling staffing.

Are College Students More Disturbed Today? Stability in the Acuity and Qualitative Character of Psychopathology of College Counseling Center Clients: 1992-1993 Through 2001-2002

Are clients at college counseling centers more disturbed today than they were 5, 10, or 20 years ago? Since the mid-1980s, counseling center personnel have consistently perceived student clients as more distressed or reported acutely distressed clients as more prevalent than they did 1, 3, or 5 years ago. Only 3 studies employing a systematic, data-based methodology have addressed this question. Two researchers used different actuarial measures of the acuity or character of client pathology and concluded that current student clients are not more disturbed than were earlier ones. One researcher used clinical judgments rendered at the end of therapy and concluded that current students were more distressed. Using the Personality Assessment Inventory, the author found the actuarially determined quantitative and qualitative indices of pathology unchanged for 3,400 counseling center clients seen during 10 consecutive years. Over the same period, the use of medications increased fivefold. An actuarial measure of client distress appears crucial to both accurate diagnosis and treatment and to the adequate preparation of future counseling center professionals.

Four eras of study of college student suicide in the United States: 1920-2004.

Studies of college student suicide can be grouped into the following 4 eras: 1920-1960, 1960-1980, 1980-1990, and 1990-2004. The suicide rate for students has declined monotonically across these 4 eras, from 13.4 to 8.0 to 7.5 and, most recently, to 6.5. The decreasing proportion of men in the student populations studied largely accounts for this decline. Since 1960, the suicide rate for students has consistently been about half the rate of the general US population, matched for age and gender. This highly favorable relative suicide rate is the result of firearms having been effectively banned from campuses. Additional population-oriented approaches warrant implementation to further reduce student-suicide rates. Approaches focused on high-risk groups also hold promise. These findings are based upon and may be most valid for the 70% of all students who attend 4-year colleges and universities full time.

Rate, Relative Risk, and Method of Suicide by Students at 4-Year Colleges and Universities in the United States, 2004-2005 through 2008-2009

A total of 622 suicides were reported among students attending 645 distinct campuses from 2004-2005 through 2008-2009. Adjusting for gender in the population at risk of 14.9 million student-years and for the source of these data, the student suicide rate of 7.0 was significantly and substantially lower than for a matched national sample. Suicide rates by firearm were significantly and substantially lower for both female and male students. Hanging was significantly and substantially lower for male students, less prominently so for female students. It is principally the ninefold decrease in the availability of firearms on campuses (vs. homes) and secondarily other features of the campus environment that are the bases for lower student suicide rates.

Comparing the Risk of Suicide of College Students with Nonstudents.

Suicide rates for college students were compared with rates for nonstudents using two definitions of students: full-time enrollment at 4-year institutions, and all-inclusive postsecondary enrollment. For traditionally aged females, comparing the suicide rate of students (3.0 per 100,000) to the rate for same-age nonstudents rather than the same-age general population did not meaningfully affect the relative risk for suicide. For traditionally aged male students (overall suicide rate 10.25 per 100,000), changing the comparison group to same-age nonstudents rather than the same-age general population did meaningfully enhance the favorable relative risk for students aged 18–24 in the smaller student cohort, and for those 18–29 in the more inclusive student cohort. These findings are interpreted as reinforcing the importance of environmental factors in affecting suicide rates, specifically the importance of restricting access to firearms in reducing rates.

Factors affecting the usage rate for mental health sections of college and university health services.

Abstract The history of the Mental Health Annual Program Survey (MHAPS) is briefly reviewed. Characteristics of the institutions participating in MHAPS during the 1975–76 program year are reported. Usage rate data for the sample are reviewed, and the procedures employed in determining the predictors of sample variability in mental health section usage rate are described. Variability in usage rate was regressed against six categories of variables: institutional characteristics, alternative available resources, mental health section staffing levels and staffing patterns, allocations of staff time, and programming options. When the significant predictors included in these six categories were pooled, five variables were sufficient to account for 95% of usage rate variance. The number of weekly interview hours per 1000 full-time students and the proportion of full-time staff each accounted for 29% of usage rate variance. Visits per patient accounted for 18%, time allocated to consultative services for 10%, and...

Is Student Pathology Really Increasing? Seven Measures of the Acuity of Counseling Center Clients, 1992–2007

Since 1992, an overwhelming consensus among counseling center directors has asserted that, year after year, the severity or psychopathology of counseling center clients has been increasing. In contrast to this perceptual consensus, the search for confirming evidence using client self-report measures has been frustrating. These studies have generally failed to confirm changes in client severity or qualitative psychopathology. The present study analyzed Personality Assessment Inventory (PAI) data from 4,755 clients spanning a 15-year period from 1992–2007. Results indicated a stable proportion of clients warranting any Axis-I diagnosis (72%), no change in overall acuity, a significant decline in suicidality, and stable levels of severity within each of four groups of clients: those warranting a PAI-assigned diagnosis for (a) a mood disorder, (b) anxiety or phobic disorder, (c) dysthymic disorder, or (d) adjustment disorder. A number of processes are identified as likely contributors to the persisting but perhaps erroneous perceptual consensus about increasing pathology.

Commentary on “Variability in College Student Suicide

ABSTRACT Stephenson, Belesis, and Balliet (2005) urge that future research on college student suicide should not continue its historical emphasis on completed suicide nor on the contrast between students and non-students. But their paper does not provide an adequately complete and accurate summary of our present knowledge of completed student suicide. This commentary, by redressing these shortcomings, provides an up to date review of our present knowledge of college student suicide. In doing so it enhances both the context for considering future research directions and the bases for the design of future studies.

Analyzing and Communicating Results

Abstract The final article in this series on research in college health deals with the analysis and communication of research data. The authors introduce descriptive and inferential statistics and summarize the applicability of different types of statistical analysis to college health research. Two major means of conveying results, conference presentations and journal articles, are also described and some guidelines for the effective communication of findings are offered.

Session Limits and the Probability of the Next Visit

ABSTRACT Among the approaches that have been used to balance the demand for services with the agency’s service resources is to limit the number of visits available to clients. Such limits have been introduced, however, without any empirical basis for estimating the impact of a limit on the overall demand for services or the proportion of clients. The probability of a next visit is a statistic that addresses these shortcomings. Analyses indicate that there is a fairly stable pattern to students use of counseling services, that this pattern is affected by the imposition of a limit, and that reducing the probabilities of a next visit for the first, second, and third visits has a greater impact on reducing total service demand than imposing a limit at the 12th visit, the most commonly reported point