Identification of Patients at Risk for Suicide and/or
Substance Abuse
|
Published: October, 2001
|
Another function of the ALERT system is to target cases with
recurrent suicidal ideation and/or high potential for substance
abuse in order to ensure that services of adequate intensity
are provided. The system incorporates information on these
risk factors from the Provider Assessment Report (PAR) and
from the LSQ/YLSQ.
In order to check the congruence between the provider assessment
and the patient self-report, an analysis was conducted on
the entire sample. The report design matched PARs with LSQs
administered within 7 days of one another. This produced 1,391
cases with concurrent PAR and LSQ data.
Of these cases, 56 (4%) reported suicidal ideation "Frequently"
or "Almost Always" on the LSQ item "I have
thoughts of ending my life." The average LSQ score for
patients reporting this level of suicidal ideation is 77 points,
which is at the high end of the severe range. In fact, a score
of 77 or higher occurs in approximately 5% of cases. These
same cases also have an average response of 3 ("Frequently")
on the item, "I feel hopeless about the future."
Over 70% of these members with recurrent suicidal thoughts
have scores of 3 or higher on the hopelessness item.
These are clearly very distressed individuals with high levels
of depression and anxiety, frequent suicidal ideation, and
a high level of hopelessness. Surprisingly, almost two thirds
of these cases were assessed by the practitioner as presenting
either with no suicidal thoughts or only a mild level of ideation.
The following graph shows the breakdown of provider assessment
of suicidal ideation for these cases.
It is not possible to offer a definitive explanation for
the discrepancy in assessments, but numerous conversations
with practitioners suggest there may be two very different
sources. First of all, there are instances when practitioners
seem genuinely surprised by the discrepancy when contacted
by phone by Care Managers, and it would seem that these practitioners
did not detect the presence of suicidal thinking. However,
the other source of discrepancy is a thorough assessment by
the practitioner who rates the patient’s suicidal thinking
as passive and "mild" in terms of risk. In other
words, the practitioner may be well aware of the patient’s
frequent suicidal thoughts, but the rating of "mild"
ideation on the PAR reflects a more comprehensive analysis
of the risk factors associated with those thoughts.
A similar analysis was performed to check the congruence
between the provider assessment and the patient self-report
regarding substance abuse. This analysis compares responses
on the LSQ chemical dependency (CD) items and the clinician
assessment of substance abuse problems on the PAR. The LSQ
has three CD-related items. These are scored on a five-point
scale, from 0 for "Never" to 4 for "Almost
Always." The CD scale score is derived by adding the
three items.
- I use alcohol or a drug to get going in the morning.
- People criticize my drinking (or drug use).
- I have trouble at work/school or other daily activities
because of drinking or drug use.
Arguably, any response other than a "Never" on
all three items could be indicative of a problem. Twenty percent
(20%) of the sample responded with at least one "Rarely."
However, its difficult to have confidence that a single rarely
is strong evidence of substance abuse problem.
For purposes of this analysis, a more stringent requirement
of a score of four or higher was used to indicate presence
of a substance abuse problem. Less than five percent (5%)
of members scored this on this scale. It appears that virtually
all patients self-reporting this degree of difficulty have
a substance abuse problem.
Similar to the case with the suicide item, a significant
percentage of these self-admitted substance abusers are not
identified by the provider as having a problem. The following
graph displays the results. Fifty eight percent (58%) of the
cases targeted by the LSQ were not identified by providers
as having a problem.
Back to Top
 
|
