How effective are interventions with caregivers? an updated meta-analysis

Silvia Sörensen, Martin Pinquart, Paul Duberstein

Research output: Contribution to journalArticlepeer-review

768 Scopus citations


Purpose: The purpose of this study was to determine the effectiveness of interventions for family caregivers of older adults. Design and Methods: Meta-analysis was used to synthesize the effects of 78 caregiver intervention studies for six outcome variables and six types of interventions. Results: The combined interventions produced a significant improvement of 0.14 to 0.41 standard deviation units, on average, for caregiver burden, depression, subjective well-being, perceived caregiver satisfaction, ability/knowledge, and care receiver symptoms. Intervention effects were larger for increasing caregivers' ability/knowledge than for caregiver burden and depression. Psychoeducational and psychotherapeutic interventions showed the most consistent short-term effects on all outcome measures. Intervention effects for dementia caregivers were smaller than those for other groups. The number of sessions, the setting, care receiver age, caregiver age, gender, type of caregiver-care receiver relationship (spouse vs adult child), initial burden, and study characteristics moderated the observed effects. Implications: Caregiver interventions are effective, but some interventions have primarily domain-specific effects rather than global effects. The differences between intervention types and moderators suggest ways of optimizing interventions.

Original languageEnglish (US)
Pages (from-to)356-372
Number of pages17
Issue number3
StatePublished - 2002
Externally publishedYes

All Science Journal Classification (ASJC) codes

  • Gerontology
  • Geriatrics and Gerontology


  • Caregiving
  • Dementia
  • Elderly
  • Meta-analysis
  • Older adults
  • Treatment

Fingerprint Dive into the research topics of 'How effective are interventions with caregivers? an updated meta-analysis'. Together they form a unique fingerprint.

Cite this