Early detection of alcohol abuse and psychiatric conditions among homeless people

Early detection of alcohol abuse and psychiatric conditions among homeless people - Homeless screener

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

NL-OMON

Registry ID

NL-OMON36242

Enrollment

400

Registered

2011-05-27

Start date

2011-10-01

Completion date

Unknown

Last updated

2024-05-06

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Alcoholism Depression Psychosis

Interventions

None listed

Eligibility

Age

18 Years to 99 Years

Inclusion criteria

Inclusion criteria: - Present in homeless care - Signed consent form

Exclusion criteria

Exclusion criteria: - Acute psychosis - Acute intoxication

Design outcomes

Primary

Measure	Time frame
Besides the items identified in phase 1, the CIDI will be used as a gold standard used to determine depression, psychotic disorders, alcohol abuse and dependence. Then, calculations are made to determine if specific items from the instrument derived from phase I correctly identify clients with a disorder (i.e. according to the CIDI diagnosis), above chance level. This is done by calculating a ROC curve (receiver operating curve). This curve indicates how well an item predicts if someone has a disorder - or not. At an Area under the Curve of 1 the prediction is perfect, while an area under the curve of 0,5 indicates chance level results. Promising items will be checked for quality (sensitivity, specificity, predictive value). In the screener of the pilot study (2007), for example, an item aimed at measuring alcohol abuse was the following: "How often do you have six (women: four) or more drinks in one day?". The response categories ranged from 1 ("never") to 5 ("daily"). The area under the curve for this question was 0.91, which indicates good quality. Subsequently, the sensitivity and specificity were calculated for various cut-off scores. A cut-off score of 3 ('1 -3 times per month ') was related to the highest sensitivity, 90%. This means that 90% of those with a diagnosis of alcohol abuse scored 3 or higher, and therefore were correctly detected. Specificity was relatively high as well at 78%. This means that 22% of the screened non-abusers were erroneously classified as abuser. The above analysis will be performed separately for alcohol abuse, for depression, and for psychotic disorders. In this way, a screener will be developed which shall preferably consist of one question per disorder. However, analyses will also check if combinations of items lead to better prediction than individual items.	—

Measure

Time frame

Besides the items identified in phase 1, the CIDI will be used as a gold standard used to determine depression, psychotic disorders, alcohol abuse and dependence. Then, calculations are made to determine if specific items from the instrument derived from phase I correctly identify clients with a disorder (i.e. according to the CIDI diagnosis), above chance level. This is done by calculating a ROC curve (receiver operating curve). This curve indicates how well an item predicts if someone has a disorder - or not. At an Area under the Curve of 1 the prediction is perfect, while an area under the curve of 0,5 indicates chance level results. Promising items will be checked for quality (sensitivity, specificity, predictive value). In the screener of the pilot study (2007), for example, an item aimed at measuring alcohol abuse was the following: "How often do you have six (women: four) or more drinks in one day?". The response categories ranged from 1 ("never") to 5 ("daily"). The area under the curve for this question was 0.91, which indicates good quality. Subsequently, the sensitivity and specificity were calculated for various cut-off scores. A cut-off score of 3 ('1 -3 times per month ') was related to the highest sensitivity, 90%. This means that 90% of those with a diagnosis of alcohol abuse scored 3 or higher, and therefore were correctly detected. Specificity was relatively high as well at 78%. This means that 22% of the screened non-abusers were erroneously classified as abuser. The above analysis will be performed separately for alcohol abuse, for depression, and for psychotic disorders. In this way, a screener will be developed which shall preferably consist of one question per disorder. However, analyses will also check if combinations of items lead to better prediction than individual items.

—

Secondary

Measure	Time frame
n.v.t.	—

Countries

Netherlands

Outcome results