Debating the Stages of Grief, Death and Dying

When researchers have a disagreement about what the research shows, most usually either submit a letter to the editor, or an editorial to the journal in question. Sometimes they’ll go one step further and even design an experiment to reproduce the effects of the previous research in question.

But rarely do they turn to a magazine to call into the question of a peer-reviewed research study. And especially not one published in the prestigious medical journal JAMA.

So you have to wonder what led Russell Friedman and John W. James to publish their treatise against the traditional and well-accepted stages of grief in the latest issue of Skeptic magazine, calling into question the results of the Yale Bereavement Study (YBS). The Yale study appeared a year and a half ago in JAMA.

Their first argument is that “stages” of grief or loss is a hypothetical concept never “proven” as fact. They note that Kübler-Ross proposed the stages of grief in her book On Death and Dying, not in a research study, which is well-known. (Kübler-Ross actually adopted Bowlby and Parkes’ theories on grief.) They delve into Kübler-Ross’s own biases which may have led her to propose the specific stages she did (and indeed, as conjecture, these are interesting observations).

But in order to frame a research investigation, one must start with certain hypotheses in order to prove them. To start one’s argument against the YBS study by suggesting that one can’t invoke hypothetical constructs is a non sequitur. (If all research started from only studying accepted facts, we’d have nothing left to study.)

So Friedman and James seem very interested in scientific facts and empirical, rigorous data. Yet they start off their article with the assertion that they are not going to argue against the YBS data with competing scientific data, but from the authors’ “having worked directly with over 100,000 grieving people during the past 30 years.” That’s a fantastic number.

In science, we call such data “anecdotal.” Because even though it’s a huge number, it’s directly colored by the two people’s pre-existing assumptions, beliefs, and their own theories about grief and loss. So while they may believe they are recounting objective data, they are doing so through their own rose-colored glasses. Research bias is such a well-understood and accepted phenomenon that most research studies conducted will specifically account for it by using third-party, independent raters or objective measures, things that are not directly involved in the research.

We’ve established, then, that these authors are not going to argue against the YBS data with scientific, comparable data, but from opinion. So the Skeptic article is one big opinion piece, masquerading as science. That helps put it into some context.

Because the Yale Bereavement Study, using actual scientific data, found strong empirical support for the five stages (which they re-label as “grief indicators”), just not in the order Kubler-Ross first hypothesized.

Although the temporal course of the absolute levels of the 5 grief indicators did not follow that proposed by the stage theory of grief, when rescaled and examined for each indicator’s peak, the data fit the hypothesized sequence exactly.

In other words, the data support the concept of there being a group of five emotions and beliefs that most people who experience grief experience to some degree and in some order. What’s the real order?

Yearning (bargaining) was the most frequent negative psychological response reported throughout the study observation period. […] Models that tested for phasic episodes of each grief indicator revealed that disbelief about the death (denial) is highest initially. As disbelief declined from the first month postloss, yearning rose until 4 months postloss and then declined. Anger over the death was fully expressed at 5 months postloss. After anger declines, severity of depressive mood peaks at approximately 6 months postloss and thereafter diminishes in intensity through 24 months postloss. Acceptance increased steadily through the study observation period ending at 24 months postloss. Because of the minuscule probability that by chance alone these 5 grief indicators would achieve their respective maximum values in the precise hypothesized sequence, these results provide at least partial support for the stage theory of grief.

The Yale researchers also decided to study stages that don’t directly or nicely fit with prior hypothesized stages. So instead of using the concept of “denial,” they felt more comfortable with the term “disbelief” to describe that stage. And yearning was substituted with the idea of “bargaining,” because it had more empirical support in the research.