A History of Prolonged Grief Disorder's Inclusion in the DSM — And What Is Missing From It

The article “Prolonged Grief Disorder and the DSM: A History” by Zachar, First, and Kendler (Journal of Nervous and Mental Disease, 2023) provides a fascinating behind-the-scenes account of how Prolonged Grief Disorder (PGD) made it into the DSM and of the debates that took place within the complicated/prolonged grief research communities and the relevant DSM committees.

Two things in particular stand out to me.

One. Those unfamiliar with the scientific research on pathological grief may be surprised to discover the degree of scientific support behind the construct of prolonged, complicated grief. The diagnostic proposals were backed by a substantial body of evidence from multiple research groups (two groups in particular, led by Holly Prigerson and Katherine Shear respectively). Both groups conceptualized the construct a bit differently, emphasized different aspects of it, relied on different sorts of samples (clinical vs community samples), and optimized the avoidance of false-negatives vs false-positives differently, but the degree of concordance and convergence was abundantly clear. The disagreements between the research groups played a central role in the DSM process, and I’ll refer you to the article to learn how the events unfolded.

More fundamentally, the naïve idea expressed ad nauseum by many popular critics of psychiatry, that DSM constructs are created out of thin air based on committee votes with little to no consideration of scientific evidence simply isn’t borne out when we look at the actual process as exemplified here. In fact, one may even argue that in this instance there is a hyper-focus on certain empirical details at the expense of other sorts of considerations. The scientific body of evidence is quite clear that there is a subset of grieving people who experience intense, complicated, enduring, and at times debilitating, forms of grief, and this grief does not respond well to standard depression treatments (antidepressants, psychotherapy aimed at depression) but does seem to improve in many cases with interventions targeting grief (psychotherapies aimed at grief).

Two. It is striking that certain concerns and debates around PGD that loom large in the public and critical discussions are almost entirely absent from this oral history of the DSM process. There is for example considerable discussion in the article of the issues around the psychometric structure of prolonged grief (whether it can be modeled as a single latent variable, e.g.) and what combination of symptoms provides the optimal diagnostic threshold, but there is practically no discussion on the nature or validity of the distinction between normal vs disordered grief, or the sense in which prolonged grief meets DSM definition of mental disorder, or worries about the ways in which the diagnostic criteria may be overapplied in different clinical settings or may lead to unwarranted consequences, etc.

How could it be? The DSM leadership was clearly not unaware of these concerns. Did these concerns simply not matter to them? Were these concerns not judged to be scientific enough to merit consideration?

In the New York Times coverage of the diagnosis, there are at least hints that such considerations played an important role:

“The most sensitive question of all was this: How long is prolonged?

Though both teams of researchers felt that they could identify the disorder six months after a bereavement, the A.P.A. “begged and pleaded” to define the syndrome more conservatively — a year after death — to avoid a public backlash, Dr. Prigerson said.”

There is no hint of this in the JNMD history at all. On the matter of finalizing the duration criterion, it simply notes:

“In their report to the Internalizing Disorders Review Committee, the expert panel recommended setting the diagnostic threshold at three criteria and keeping the DSM-5 duration cutoffs, which for adults was at least 12 months since the death, but for children, 6 months. Evidence that clinicians did not favor the 6-month duration in the ICD played a role in this decision (Dietl et al., 2018).”

Was the begging and pleading sanitized away from the JNMD history, or was it simply Prigerson’s interpretation of events to begin with? I don’t know.

Maarten C Eisma has discussed many of the controversies around PGD in an article for the Australian & New Zealand Journal of Psychiatry, noting that the unclear distinction between normal grief and PGD poses a major challenge to the validity and that concerns have been raised about potential negative societal consequences in the form of medicalization of normal grief, risks of novel pharmacotherapies and stigmatization of people diagnosed with PGD.

The general concerns around PGD are expressed well by Cacciatore and Frances in their letters to Lancet Psychiatry, arguing “Mental disorders that might work well in the hands of researchers are often a disaster when applied in general practice.” [link] and “We agree that a subset of grieving people experience intense and enduring symptoms of grief. However, our concerns remain that certain types of losses pose a substantially higher risk of misdiagnosis, and that context will be routinely dismissed, leading to diagnostic inflation. We agree that some grieving people will need extra support, although primarily because of the circumstances of the death and the closeness of the relationship, not because of pathology, and thus we disagree that a diagnosis is required in these situations.” [link]

They also stated: “Although some assert that only a small percentage of grievers will meet the criteria for prolonged grief disorder, evidence suggests that the proportion might be much higher under certain circumstances.”

Prigerson had estimated that the DSM-5 prolonged grief disorder criteria would apply to about 4% of the bereaved individuals. However, a cross-sectional survey study by Thieleman, Cacciatore and Frances, published a few weeks ago, assessing PGD rates using the Prolonged Grief-13-Revised (PG-13-R) among a large sample (n = 1137) of bereaved individuals, reported that 34.3 % of the sample met PGD criteria. Vast majority of the respondents viewed the relevant symptoms as normal. One could argue that a diagnosis based on PG-13-R isn’t quite equivalent to a DSM diagnosis of PGD, and that’s true, but this is the sort of strategy that is near-universally applied to estimate and identify cases of psychiatric disorders such as depressive and anxiety disorders. It is easy to appreciate that initial conservative estimates almost always give way to estimates that are many folds higher in practice.

The Thieleman et al. paper had important things to report about what the respondents thought about the diagnosis as well:

“When asked, “Overall, do you feel your emotional responses, as painful as they may be, to this person's death are understandable and normal?,” 98.1 % of respondents answered “yes.” Respondents were also asked, “If you were told by a professional that your grief was a mental disorder, how helpful or unhelpful would you find this?” Respondents generally viewed the idea that their grief was a disorder unfavorably, with 54.4 % choosing very unhelpful, 7.9 % choosing somewhat unhelpful, 25.9 % indicating they weren't sure, 7.6 % choosing somewhat helpful, and 4.2 % choosing very helpful.”

Concerns about medicalization are also notable in the coverage of the diagnosis by the New York Times:

“Filipp Brunshteyn, whose 3-year-old daughter died after an automobile accident in 2016, said grieving people could be set back by the message that their response was dysfunctional.

“Anything we inject into this journey that says, ‘that’s not normal,’ that could cause more harm than good,” he said. “You are already dealing with someone very vulnerable, and they need validation.””

Cacciatore was also quoted:

“When someone who is a quote-unquote expert tells us we are disordered and we are feeling very vulnerable and feeling overwhelmed, we no longer trust ourselves and our emotions,” Dr. Cacciatore said. “To me, that is an incredibly dangerous move, and short sighted.”

There is a lot more awareness now that psychiatric diagnosis — judgments of pathology and disorderedness and the assumptions about categorization and intrinsic essences as understood by the patients — can potentially alter the natural history of the symptoms and the behavior of the patients in negative ways. (“Mental illnesses are often seen as chronic and intractable forces that take over our lives, but I wonder how much the stories we tell about them, especially in the beginning, can shape their course. People can feel freed by these stories, but they can also get stuck in them.” Rachel Aviv)

Lars Petter Sødal Bergsmark and Frida Ramsing have an excellent paper “Which Considerations Are Lost When Debating the Prolonged Grief Disorder Diagnosis” in Theory & Psychology that I really like. They write:

“We perceive that the implementation of prolonged grief disorder as a diagnosis is controversial and provocative to a lot of scholars for two main reasons. The first being an existential or philosophical question pertaining to what extent grief could and should be understood as a disorder requiring treatment (e.g., Brinkmann, 2018a). This point is also the one most commonly expressed outside scholarly discussion in the general public. There seems to be something unnatural about perceiving the reaction to losing a loved one as a medical condition requiring treatment.1 Secondly, the implementation of “the grief diagnosis” raises the larger scientific discussion of “diagnostic cultures” (Brinkmann, 2016) and “psychologicalization of society” (Madsen, 2018). Those authors contend that more and more aspects of human existence are becoming the subject of diagnostics, and are handled by medical authorities.

The pronounced disagreement about grief between cultural psychologists and health psychologists is detrimental to the scientific dialogue (see Johansen et al., 2021), and furthermore, we believe, to the public’s comprehension of the rationale and actual limitations of psychiatry. An important notion is that the PGD diagnosis is aimed at a group of people who, in definite terms, experience a significant and debilitating degree of suffering and that universal human grief does not fit this description. Naturally, there are some challenges to this distinction (which we’ll see later), but when discussing clinical or pathological grief, two distinct groups will emerge (the patient population and the general public). This distinction is often overlooked or under-communicated.”

“This article centers around two main arguments. First, we hope to shed light on why the patient population, which the diagnosis is aimed at, does not exhibit common grief experiences, but namely pathological grief reactions. We are concerned that uncritical, existential opposition against the grief diagnosis risks framing the concept of grief in a way that creates larger distances between the two academic perspectives. Second, we address why diagnoses’ current societal function is problematic and worth evaluating, especially if presented as neutral and apolitical. This paper thus outlines the need for cultural psychological insights in clinical practice. In this regard, the implementation of PGD should definitely generate a critique of diagnoses leveled at the overall psychiatric handling of mental suffering, including the diagnoses’ various functions. Crucially, this is not a problem restricted to grief diagnosing, and should neither be formulated as such.”

“To summarize, our analysis indicates that the implementation of PGD raises very few grief-specific issues, but illustrates some pervasive and serious problems with psychiatric diagnostics as a whole.”

I think Bergsmark and Ramsing are absolutely correct in this analysis. Issues of medicalization, self-understanding, misuse, etc. are not specific to prolonged grief. But they certainly apply to prolonged grief, and one would hope that people with the authority to create new official diagnoses would spend at least some time thinking about how these issues apply to a new diagnosis.

Sometimes a diagnostic construct can be helpful in a specialized setting: as a prototype, it can capture a particular sort of presentation that is of clinical importance, that we need to be aware of and address when we encounter it. If we take the same construct, operationalize it into simple descriptions that apply across settings, and actively screen for it in unsuspecting individuals, we will end up with a situation where that prototype is over-applied, where much of its original utility has been diluted, and where, perhaps, it even harms more than it helps.

Overall, it’s a good thing to identify syndromic patterns of distress and disability, and to make help available to those who experience them. It starts becoming a problem when these patterns acquire a scientific authority that exceeds their reality, when these patterns, once operationalized into XYZ criteria and straddled with assumptions of dysfunction and pathology, acquire such epistemic power that one’s experiences and parts of one’s self are now interpreted through those patterns, whether one wants to or not, and when one can no longer seek care/help/treatment/accommodations without subjecting oneself to these assumptions.

I absolutely think that prolonged/complicated grief is a recognizable clinical entity that can be very disabling and that it is a valid target of therapeutic efforts, and that it is deserving of clinical care and accommodations. But it speaks to the complicated, even corrupting, influence of contemporary nosologies, and the logic of dysfunction, categorization, and operationalization, that making official recognition available to grieving individuals also makes them vulnerable to epistemic harm and medical exploitation.

Since “dysfunction” is an essential feature of the DSM concept of mental disorder, one would expect some conceptual and scientific discussion that “prolonged grief” actually reflects “a dysfunction in the psychological, biological, or developmental processes.” Such a discussion would require DSM to explicitly state what dysfunction means (presently left undefined) and how its presence can be demonstrated or debated in clinical/scientific contexts, and how disputes can be resolved. If the notion is common-sensical or folk psychological, which I think it is and which others such as Kendler have maintained, surely this warrants an explicit acknowledgment to the public given the tendency to interpret dysfunction in naturalist terms?

According to DSM-5-TR: “Prolonged grief disorder represents a prolonged maladaptive grief reaction… Prolonged grief disorder is distinguished from normal grief by the presence of severe grief reactions that persist at least 12 months (6 months in children or adolescents) after the death of a person who was close to the bereaved individual. It is only when severe levels of grief response persist for the specified duration following the death, interfere with the individual’s capacity to function, and exceed cultural, social, or religious norms that prolonged grief disorder is diagnosed.” It is part of the diagnostic criteria for PGD that “The duration and severity of the bereavement reaction clearly exceed expected social, cultural, or religious norms for the individual’s culture and context.”

New York Times ran their story on the front page on March 19, 2022 with the heading “Prolonged Grief Is Declared a Mental Disorder.” For the public, the story, the controversy, the protest, is about the mental disorder part. It is not about the fact that grief can be prolonged or complicated, but that it can, in some authoritative medical sense, be declared a mental disorder. And yet, it is precisely this aspect of the story that barely gets any attention in the DSM internal process.

The explanations offered here and there for PGD being a disorder are all over the place.

According to Psychiatric News (Nov 25, 2020): “The diagnosis of prolonged grief disorder is based on a large body of research indicating that some people experience persistent difficulties associated with bereavement that are substantially prolonged beyond culturally normative expectations, Kenneth Kendler, M.D., vice chair of the DSM Steering Committee, told the Board in October.” In the same Psychiatric News piece, Kendler is quoted as saying: “There is substantial evidence that the proposed category would meet criteria for a mental disorder…” But what constitutes substantial evidence that PGD meets criteria for mental disorder? The fact that there are persistent difficulties that are prolonged beyond culturally normative expectations?

In a 2009 publication, Prigerson et al say: “Our results indicate that PGD meets DSM criteria for inclusion as a distinct mental disorder on the grounds that it is a clinically significant form of psychological distress associated with substantial disability.”

Prigerson and Maciejewski wrote in their reply to Cacciatore and Frances: “Because a diagnosis of prolonged grief disorder requires grief-related distress and functional impairments, and because it is significantly associated with concurrent and future depression and suicidality, and also a risk for other adverse health outcomes (eg, bodily pain, worse general health, vitality, social and role functioning), this diagnosis has, by definition and by data, proven indicative of a pathological response to bereavement.” (my emphasis)

Well, the DSM definition requires a “dysfunction” and I’m not seeing anyone clearly acknowledge what the dysfunction here is.

To be clear, I am not saying that there is no dysfunction in PGD. I think there is a dysfunction in a very folk psychological, common sensical, value-laden kinda way; it’s this notion that something (the process of grieving, in this case) is not functioning as it is supposed to, but the norms for “as it is supposed to” invoked here are basically personal and sociocultural norms. If that is true, then characterizing prolonged grief as a dysfunction amounts to little more than: the diagnostic manual now officially recognizes that some people experience persistent difficulties associated with bereavement that are intense and substantially prolonged beyond socioculturally normative expectations. Obviously, for a lot of people, including many philosophers and psychiatrists, calling something a mental dysfunction is a much stronger claim that says something about failure of natural design, or neurophysiological alterations, etc. For them, it’s not just: “Oh, you are suffering in a way that is socioculturally atypical and we have clinical interventions that can potentially help you.” Which is what the controversy is all about (see).

Back to my puzzlement… why is all this missing in the actual scientific debates taking place in the DSM committees? It’s just odd that the biggest concerns that people have about this diagnosis — Why is it a disorder? How can the diagnosis be misused in clinical settings? How will it negatively impact people’s self-understanding? Etc. — are the ones generally missing from the deliberations of the DSM committee, at least in the history of it as reported by Zachar, et al. They obviously can’t make stuff up if it didn’t happen. If the major players in the incorporation of PGD in the DSM were indifferent to these issues, the history is what it is. But the indifference does tell us something important about the DSM process.

---

P.S. Many people misunderstand the nature of the time duration requirement for PGD. Consider, for instance, Martha Weinman Lear’s letter to the editor of the New York Times: “I have had grief (longer than a year), have interviewed scores of people in grief, and have written about grief, and what strikes me as abnormal is not grief prolonged beyond the A.P.A.’s one-year prescription, but the degree of chutzpah required, professional training notwithstanding, to presume to set timelines for the normal grief of others, which in fact is as various as the grievers themselves.”

The DSM, in fact, recognizes that grief isn’t automatically a disorder after one year and doesn’t set any rigid timelines for normal grief. DSM-5-TR notes: “Although in general this time frame reliably discriminates normal grief from grief that continues to be severe and impairing, the duration of adaptive grief may vary individually and cross-culturally.” The one-year requirement is also necessary, but not sufficient. Other conditions need to be met, including symptom severity, clinical significance, and deviation from sociocultural norms. The only thing the one-year requirement says is that it can’t be PGD before one year. It doesn’t mean that anything after one year is abnormal.

---

See also: