Depression diagnoses and depression algorithm
Contact: Almar Kok
Depressive disorders are common in older adults; the 1-year prevalence of major depressive disorder (MDD – referred to in the remainder of this documentation as ‘depression’) among older adults (>65 years) ranges from 1-5% (Fiske et al., 2009). Depression can have serious consequences for quality of life and lead to high health care costs. Depression often has an early onset and a chronic or remittent course into later life. However, also a large number of people experience a first episode of depression in later life. Comorbidity of depression with physical diseases and cognitive impairment is high in older adults. To investigate the burden of depression in later life, LASA includes a screening instrument for depression (Center for Epidemiologic Studies-Depression scale; CES-D) and a diagnostic interview for those scoring high on the screener. Additionally, LASA collects data on drug use, including antidepressants, and has data from participants’ general practitioners on the presence of depression, and, since the K-wave, lifetime presence of depression and other psychiatric disorders.
Measurement instruments in LASA
Depression diagnoses are obtained according to a two stage procedure. At baseline, all persons with a CES-D > 16 and a similarly sized random sample of the non-depressed (CES-D < 16) were approached to participate in the diagnostic interview, which was part of the medical interview (Beekman et 1997). At each follow-up measurement, all persons with a CES-D > 16 at the current and previous measurement are invited to participate in the diagnostic interview. Since wave G, persons were also invited for the diagnostic interview when having a HADS-A score > 8 (anxiety screening).
In wave B through 3B, the Diagnostic Interview Schedule (DIS, Robins et al. 1981; Dingemans et al 1985) was used, which was based on DSM III criteria. From wave I onwards, the CIDI (Composite International Diagnostic Interview) was used to assess disorders according to DSM IV criteria. The DIS and the CIDI are designed for epidemiological research and have been widely used in older populations. Interviewers were fully trained by certified staff, using the official Dutch translation. Diagnoses include Major Depressive Disorder (MDD) and dysthymia. Age of onset and recency are also assessed.
In addition to the CES-D and DIS/CIDI interviews, there are more data sources available in LASA that may indicate the presence of a depressive disorder, such as data from general practitioners about the presence of depression and information about the use of antidepressants. To make optimal use of all data sources and complement missing data on one data source with data from another, we combined them into an algorithm. Based on data from the CES-D, DIS/CIDI, General Practitioner data and use of antidepressant medication, the algorithm describes the likelihood that a participant had a depressive episode at each of the LASA waves. This means: whether depression was present between the previous and current wave. The algorithm only covers MDD and not dysthymia.
The algorithm currently spans data from waves B to I:
|LASA files used||Comments|
|CES-D ≥16||B – I
|Data from main interview and telephone with
respondent (wave C,D,E,F,G,H,I)
|DIS or CIDI||B – I||D10||Only available if participant scored ≥16 on the CES-D (except for a subsample at wave B)|
|General Practitioner (GP) Diagnosis||GP report||B, C, E, H
|G01||Diagnosis was determined with or without a medical
specialist. Diagnosis was set at missing if year of
diagnosis was not reported.
|Use of antidepressants||B – I
Algorithm rules and computation
At each wave, the status of each of the four depression indicators could be one of the following:
- Negative (not indicating depression)
- Positive (indicating depression)
- Possible (this applied to GP diagnosis ‘yes’ but without information about a year of diagnosis)
- Not applicable (for example if no CIDI diagnosis was available because participant scored <16 on the CES-D questionnaire)
The algorithm distinguishes 48 possible combinations of the four indicators. (Note: in the data-files these are numbered 0-35 and 47-58. This is because an earlier version of the algorithm included 11 additional categories that were redundant in the present version of the algorithm)
Finally, these 48 categories were collapsed and classified into five levels of likelihood of MDD:
- No depression
- Unlikely depression
- Possible depression
- Likely depression
- Very likely depression
Additionally, a category “unsure” was distinguished for situations where GP was positive but no year was known and the rest was missing or a single other available indicator was negative.
The file ZDEP contains this categorization for each of the waves B to I. An overview of the combinations of indicators leading to the categorization can be found here.
LASA*d10 files contain the raw items of the DIS, negative life events during childhood, information about treatment and family history of mental disorders.
LASA*d11 files contain the depression diagnoses per timeframe.
LASAZDEP contains categorization of the likelihood of depression based on the algorithm for waves B-I.
Availability of information per wave ¹
¹ More information about the LASA data collection waves is available here
* 2B=baseline second cohort;
3B=baseline third cohort;
MB=migrants: baseline first cohort;
Me=data collected in medical interview
Previous use in LASA
Studies have been performed on the prevalence of depressive disorders (Beekman et al. 1995a), it’s comorbidity with anxiety disorders (Beekman et al. 2000) and its risk factors (Braam et al. 1997, Beekman et al. 2000; Steunenberg et al. 2010).
Other studies focussed on the consequences of depressive disorders (Beekman et al. 1997) and course of depression (Beekman et al. 1995b) was examined. In addition, Penninx et al. (1999) studied minor and major depression and the risk of death. Sonnenberg et al. (2003, 2008) examined drug treatment and trends in antidepressant use in depressed older adults. Recently, Jeuring et al. examined the long-term outcome of subthreshold depression in later life, with one of the outcomes being major depression (in press).
- Beekman AT, Deeg DJ, van Tilburg T, Smit JH, Hooijer C, van Tilburg W. Major and minor depression in later life: a study of prevalence and risk factors. J Affect Disord. 1995a, 24;36(1-2):65-75.
- Beekman AT, Deeg DJ, Smit JH, van Tilburg W. Predicting the course of depression in the older population: results from a community-based study in The Netherlands. J Affect Disord. 1995b, 16;34(1):41-49.
- Beekman AT, Deeg DJ, Braam AW, Smit JH, Van Tilburg W. Consequences of major and minor depression in later life: a study of disability, well-being and service utilization. Psychol Med. 1997, 27(6):1397-1409.
- Beekman AT, de Beurs E, van Balkom AJ, Deeg DJ, van Dyck R, van Tilburg W. Anxiety and depression in later life: Co-occurrence and communality of risk factors. Am J Psychiatry. 2000, 157(1):89-95.
- Braam AW, Beekman AT, Deeg DJ, Smit JH, van Tilburg W. Religiosity as a protective or prognostic factor of depression in later life: Results from a community survey in The Netherlands. Acta Psychiatr Scand. 1997, 96(3):199-205.
- Dingemans P, Van Engeland H, Dijkhuis JH & Bleeker JH. De ‘Diagnostic Interview Scale’(DIS). Tijdschrift voor Psychiatrie, 1985, 27 (5): 341-359.
- Fiske A, Loebach, Wetherell J, Gatz M. Depression in older adults. Annu Rev Clin Psychol 2009; 5:363- 389.
- Hovens, J.E., Bramsen, I., Van der Ploeg, H.M., 2000. Zelfinventarisatielijst Posttraumatische Stressstoornis ZIL Handleiding. Swets Test Publishers, Lisse.
- Jeuring HW, Huisman M, Comijs HC, Stek ML, Beekman ATF. The Long-Term Outcome of Subthreshold Depression in Later Life. Psychological Medicine, in press.
- Penninx BW, Geerlings SW, Deeg DJ, van Eijk JT, van Tilburg W, Beekman AT. Minor and major depression and the risk of death in older persons. Arch Gen Psychiatry. 1999, 56(10):889-895.
- Robins LN, Helzer JE, Croughan JL, Ratcliff KS. National Institute of Mental Health Diagnostic Interview Schedule: its history, characteristics and validity. Arch Gen Psychiatry 1981; 38(4): 381–389.
- Sonnenberg CM, Deeg DJ, Comijs HC, van Tilburg W, Beekman AT. Trends in antidepressant use in the older population: results from the LASA-study over a period of 10 years. J Affect Disord. 2008, 111(2-3):299-305.
- Sonnenberg CM, Beekman AT, Deeg DJ, Van Tilburg W. Drug treatment in depressed elderly in the Dutch community. Int J Geriatr Psychiatry. 2003, 18(2):99-104.
- Steunenberg B, Beekman AT, Deeg DJ, Kerkhof AJ. Personality predicts recurrence of late-life depression. J Affect Disord. 2010, 123(1-3):164-172.
Date of last update: August, 2023