Methodological issues in studies of the effectiveness of grief intervention: Can the type of control group make a difference in assessment?

Abstract

There is as yet little consensus about the efficacy of grief interventions and a substantial number of grief intervention efficacy studies suffer from methodological shortcomings. An important aspect in efficacy studies is inclusion of a control group. The purpose of the present study is to answer the following research question: What is the expected effect of the Bereaved Families of Ontario (BFO)-intervention, given the level of impact found in other studies of bereavement intervention using comparable control groups? The non-intervention control group of the (BFO)-projects was compared to nineteen grief intervention efficacy studies. The latter were categorized into subgroups according to inclusion of four types of control groups; non-intervention, waiting list, other treatment and other. Weighted mean effect sizes were calculated using grief and depressive symptoms scales for each category and the expected effect of the BFO-intervention was estimated. It was predicted that, although the BFO-project incorporated some methodological improvements, the effect size would be small to moderate. A mixed between-within subject analysis of variance was conducted to assess the impact of the two conditions (intervention N =52, control N= 50) on participants’ scores on the HGRC (grief questionnaire) and the depression scale of the SCL-90, across three time periods (pre-intervention, post-intervention and 6 months follow-up). As was expected considering the effect sizes of the comparable grief intervention efficacy studies, the BFO-intervention effect size was small. Implications of these results are discussed.