Does it make sense to do repeated surveys?--the Lundby Study, 1947-1997

Area of Science:

• Psychiatry
• Epidemiology
• Longitudinal Studies

Background:

• The Lundby Study is a long-term epidemiological survey initiated in 1947.
• It investigates mental health trends in a defined Swedish rural population.

Purpose of the Study:

• To detail the methodology and challenges of the Lundby Study's repeated surveys.
• To assess the feasibility of longitudinal psychiatric research with evolving diagnostic criteria.

Main Methods:

• Utilized best-estimate consensus diagnoses, DSM-IV, and ICD-10 criteria across survey waves.
• Employed multiple data sources including interviews, registers, case notes, and key informants to ensure data completeness.
• Managed attrition through comprehensive data collection strategies, including posthumous data retrieval.

Main Results:

• The study successfully followed 3563 probands across multiple decades (1947, 1957, 1972, 1997-2000).
• Achieved high data completeness (98-99% in initial waves, 94% in the fourth wave despite 36% mortality).
• Observed attrition rates of 13% for interviews, with mortality data compensating for lost participants.

Conclusions:

• Low attrition rates and consistent diagnostic approaches over 50 years support the validity of temporal comparisons.
• The study highlights the importance of multiple data sources in longitudinal research to overcome diagnostic system changes.