Prolonged Grief Disorder: Misconceptions about Misconceptions. A Reply to Eisma

We appreciate Eisma’s critique of our article¹ and the opportunity to address two fundamental problems with its assertions: 1) a misunderstanding of the distinction between (mental) disorder prevalence and (psycho) pathology, and 2) overconfidence in the validity of thought experiments.

Eisma asserts that “odds of meeting PGD criteria is 24 times higher in parents of deceased children than in people who lost a parent. Therefore, for some bereaved subgroups PGD is a common (i.e., typical) response.” ² Absent from discussion is information on the prevalence of Prolonged Grief Disorder (PGD) among bereaved parents; rates needed to determine its commonality. Military veterans exposed to combat have high rates (28%³) of posttraumatic stress disorder (PTSD). Would Eisma claim that PTSD is not a psychopathological state among combat veterans? High rates of PGD among certain at-risk subgroups are alone not evidence that a PGD diagnosis pathologizes a normal bereavement response.

Eisma goes on to discount statistics from a study in which 96% of bereaved respondents reported, if diagnosed with PGD, they would be relieved to know they had a recognizable problem.⁴ He asserts that these results are of “questionable validity” because they come from a “small cross-sectional study of American widowed adults.” However, these results come from a stigma-focused sub-study embedded within a larger National Institute of Mental Health-funded prospective cohort study. By contrast, he claims that “vignette-based experimental studies,” in which respondents from the public are asked how they would react to hypothetical PGD cases, provide stronger evidence that people diagnosed with PGD will suffer from stigmatization. Concerns have been raised about vignette-based methodologies because “textual descriptions and hypothetical behavior might not be sufficiently representative of real-world phenomena, and this raises concern about the validity of research findings and conclusions based on them.”^{5 (p.161)}

Prevalence does not connote pathology and data from thought experiments should not be considered more credible than data from interviews with bereaved individuals.