What has been learned about complicated grief (CG), as distinct from normal grieving, has been gleaned from mixed-age samples to date. This communication will focus on the presentation of CG in the older subgroup after reviewing some features of CG that apply to all grieving adults. The case vignette of Sophia will illustrate the presentation of CG and its successful treatment with a new specialized treatment called complicated grief therapy or CGT. Finally, new research findings and the combined use of medication will also be addressed.

The annual incidence of spousal loss is 1.6% for men and 3% for women, resulting in over 800 000 new widows and widowers each year in the United States.1 Although the terms are often used interchangeably, bereavement refers to the state of having lost someone emotionally important (literally meaning “robbed of something valuable”) whereas grief is an instinctual response to bereavement that includes the person's “symptoms,” thoughts, feelings, and behaviors. Seventy percent of bereaved people will cope adaptively with the pain of their loss and the restorative process to a new state of function without their lost loved one, either by virtue of their own innate coping ability or in addition to the support provided to them by family, friends, and/or spiritual leaders. Thirty percent of grievers will face a complication such as major depression (15%), post-traumatic stress disorder-PTSD (depending upon the circumstances of the death), or complicated grief (10% to 20%).2 Major depression secondary to bereavement and CG are often comorbid, but each can also exist without the other.3

The natural course of grieving in late life

When comparing grievers, the only discernable pattern is the variability in intensity and course of grief over time with periods of relative quietude as well as periods of reactivation, like the ebb and flow of the tides. Periods of renewed intensity are often triggered by painful reminders, such as anniversary dates.

The twin concepts of acute grief and integrated grief are very useful in differentiating adaptive from complicated grievers. To assess this in the context of the immediate aftermath of the death of an important relationship, Eric Lindeman catalogued the reactions of a large group of surviving grievers of victims of the Coconut Grove Fire that killed 500 people in Boston in 1942.4 He found the similarities among grievers, outlined in Table I, that have stood the test of time in subsequent research, and these symptoms are now referred to as acute grief. The intensity of this phase can be affected by: the age of the victim; the suddenness of the death or the chronicity of illness leading up to the death; the quality of the relationship between the deceased and the griever; any past psychiatric history in the griever (particularly anxiety, depression, or substance abuse); coping style and the adequacy of support systems. The intensity of acute grief can be quite debilitating in the short run with social isolation, inability to work effectively, withdrawal of participation in hobbies or prior activities that brought pleasure and periods of guilty rumination, profound sadness, and frequent crying spells.

Types of grief.

Normal grief
Common symptoms of acute grief that are within normal limits within the first 6 to 12 months after the loss
1. Recurrent, strong feelings of yearning, wanting very much to be reunited with the person who died; possibly even a wish to die in order to be with deceased loved one
2. Pangs of deep sadness or remorse, episodes of crying or sobbing, typically interspersed with periods of respite and even positive emotions
3. Steady stream of thoughts or images of deceased, may be vivid or even entail hallucinatory experiences of seeing or hearing deceased person
4. Struggle to accept the reality of the death, wishing to protest against it; there may be some feelings of bitterness or anger about the death
5. Somatic distress, eg, uncontrollable sighing, digestive symptoms, loss of appetite, dry mouth, feelings of hollowness, sleep disturbance, fatigue, exhaustion or weakness, restlessness, aimless activity, difficulty initiating or maintaining organized activities, and altered sensorium
6. Feeling disconnected from the world or other people, indifferent, not interested or irritable with others
Symptoms of integrated grief that are within normal limits
1. Sense of having adjusted to the loss
2. Interest and sense of purpose, ability to function, and capacity for joy and satisfaction are restored
3. Feelings of emotional loneliness may persist
4. Feelings of sadness and longing tend to be in the background but still present
5. Thoughts and memories of the deceased person accessible and bittersweet but no longer dominate the mind
6. Occasional hallucinatory experiences of the deceased may occur
7. Surges of grief in response to calendar days or other periodic reminders of the toss may occur
Complicated grief
1. Persistent intense symptoms of acute grief
2. The presence of thoughts, feelings, or behaviors reflecting excessive or detracting concerns about the circumstances or consequences of the death

With the passage of time, and the support and encouragement of concerned family and friends, restoration to pre-death functioning levels is the rule rather than the exception within approximately 6 months after the death occurred. At this point, pangs of pain, longing, and sadness can still exist but they are more fleeting and are no longer “center stage” but rather “on the back burner.” More attention is turned to the business of getting on with life and attending to responsibilities and to the needs of others and even the bereaved person's own needs (such as medical care) all of which were temporarily neglected during the throes of acute grief. This stage is referred to as integrated grief where pain, longing, and sadness are accessible when time permits to reflect upon them but are not regularly intrusive or dominant, as is the case in CG.

Comparing acute grief, integrated grief, and complicated grief

Various labels have been used to describe pathologic variations of grieving such as chronic, delayed, and traumatic. For our purposes, we will differentiate only three terms: acute grief, integrated grief, and CG.

Acute grief characterizes the early stage of grief that include a range of emotions including shock, disbelief, sadness, anger, hostility, insomnia, and the loss of ability to function as usual. Integrated grief is a permanent state in which the griever is changed forever by the loss, but adaptation or restoration is taking place and it is the dominant activity by roughly 6 months after the loss, that is, the restoration process is predominant compared with intense yearning, reveries about the lost person, and social withdrawal. CG, in contrast, is a state of being in which the griever remains preoccupied with reminders of the reality of their loss that are persistent, severe, and pervasive, giving the griever a sense of being stuck in their grief beyond 6 months and sometimes for decades after the death has occurred.

In DSM-TV-TR,5 uncomplicated bereavement is a “V” code and there are no current designations for more complicated grief. Shear and colleagues have proposed operationalized definitions to distinguish the normal acute grief symptoms (within 6 months of the death), integrated grief (6 months or later after a death has occurred), and CG as outlined in Table I.6

Proposed criteria for complicated grief.

A. The person has been bereaved, ie, experienced the dead of a loved one, for at least 6 months
B. At least one of the following symptoms of persistent intense acute grief has been present for a period longer than is expected by others in the person's social or cultural environment
1. Persistent intense yearning or longing for the person who died
2. Frequent intense feelings of loneliness or like life is empty or meaningless without the person who died
3. Recurrent thoughts that it is unfair, meaningless, or unbearable to have to live when a loved one has died, or a recurrent urge to die in order to find or to join the deceased
4. Frequent preoccupying thoughts about the person who died, eg, thoughts or images of the person intrude on usual activities or interfere with functioning
C. At least two of the following symptoms are present for at least 1 month.
1. Frequent troubling rumination about circumstances or consequences of the death, eg, concerns about how or why the person died, or about not being able to manage without their loved one, thoughts of having let the deceased person down, etc
2. Recurrent feeling of disbelief or inability to accept the death, as if the person cannot believe or accept that their loved one is really gone
3. Persistent feeling of being shocked, stunned, dazed, or emotionally numb since the death
4. Recurrent feelings of anger or bitterness related to the death
5. Persistent difficulty trusting or caring about other people or feeling intensely envious of others who have not experienced a similar loss
6. Frequently experiencing pain or other symptoms that the deceased person had, or hearing the voice of or seeing the deceased person
7. Experiencing intense emotional or physiological reactivity to memories of the person who died or to reminders of the loss
8. Change in behavior due to excessive avoidance or the opposite, excessive proximity seeking, eg, refraining from going places, doing things, or having contact with things that are reminders of the loss, or feeling drawn to reminders of the person, such as wanting to see, touch, hear, or smell things to feel close to the person who died (Note: sometimes people experience both of these seemingly contradictory symptoms.)
D. The duration of symptoms and impairment is at least 1 month
E. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning, where impairment is not better explained as a culturally appropriate response

To diagnose CG for research purposes, Shear and colleagues use the Inventory of Complicated Grief (ICG), a validated 19-item scale7 in which a score of 30 our higher is defined as the cut-off for inclusion. A screening tool known as the “Brief Grief Questionnaire8” is shown in Table III. It contains only five questions designed to be a self-report answered according to a three point scale of frequency. A score of five or greater indicates a possible CG diagnosis to be followed with the ICG for definitive inclusion.

Brief Grief Questionaire for screening for Complicated Grief. This copyrighted instrument is reprinted with permission from Katherine Shear, MD.

If 6 months or more have passed since the death of.............., please answer the following questions:
How much of the Tone are you having trouble accepting the death of a loved one?012
How much does your grief still interfere with your life?012
How much are you having images or thoughts of your loved one when she died or other thoughts about the death that really bottler you?012
Are there things you used to do when your loved one was alive that you don't feel comfortable doing anymore, that you avoid? Like going somewhere you went with him/her, or doing things you used to enjoy together?012
Or avoiding locking at pictures or talking about your loved one? How much are you avoiding these things?012
How much are you feeling cut off or distant from other people since your loved one died, even people you used to be close to the family or friend?012

In later life, losses commonly include spouses, siblings, and peers, and less commonly adult children and grand children, the latter of which are generally perceived as unnatural and unfair. Spousal loss may have followed a lengthy illness in which the care-giving burden may have been exhausting and the death a welcomed relief from terrible suffering but also might be a possible nidus for guilty rumination over admitting to having thoughts that they sometimes hoped for the death to occur in order to relieve their burden of care, thoughts they now consider be overly selfish in retrospect.

Couples who have aged successfully as a unit of shared responsibility for each other sometimes find that being alone without their partner's contribution can feel overwhelming, with either the perception they can no longer cope in the same way and are therefore forced to accept help from other sources or possibly a move to an alternative living arrangement; or alternatively, the perception that their life is now empty without the caregiving role that organized their whole daily routine.

A marital relationship that is mutually satisfying over a long period is viewed by some grievers with gratitude for lasting as long as it did. Finding new sources of support, however, can run the gamut from greater reliance on children, friends, religious institutions, or supportive environs such as life-care communities as well as hired help and potential new romantic relationships. Remarriage is protective against depression for widowers, but not necessarily widows, and older survivors of losses through death sometimes conclude that too many complications would arise from remarriage as the multiple allegiances to each respective extended family are too cumbersome. A sustained romantic friendship is often chosen over remarriage or cohabitation in these cases.

Increasing medical burden, disability, and cognitive impairment can occur within a large age range but all are more likely to accumulate with advancing age and thus the necessity of greater dependence on others can complicate the grieving process and further lead to demoralization or depression.

Case vignette: recognizing and treating complicated grief in late life

CG and its treatment are best illustrated with the case vignette of Sophia, age 72, whose husband of 52 years had died 9 months earlier after an acute illness. He had been well up to 2 days prior to death, and they had imminent travel plans until the point his illness turned grave with complications. Sophia was shocked that her husband, a prominent local businessman, had died so suddenly. She said she felt like “some had cut her in two with a saw.”

In the 6 months after her husband's death, Sophia stopped going to church and her gym, lost 20 pounds and was noted by family members to just sit in her chair for multiple hours at a time without talking or watching television or listening to the radio. Her family grew concerned and her son moved into her home with her but continued to commute to his job. He tried to offer her some companionship and to see that she ate nutritious meals.

Sophia admitted that she had thought of suicide and that it was comforting fantasy to just drive into her garage and leave the engine running. She saw an ad for our grief study in her local paper and called asking for help.

Sophia was fully evaluated scoring 35 on the ICG endorsing intense longing, inability to accept her husband's death, feeling disbelief and being drawn to places they spent time together but also avoiding reminders. This last point was one of the most difficult things as, living in a small community, everyone knew her husband and every place she might go reminded her of going there with her husband. She also scored 13 on the Quick Inventory of Depressive Symptomatology (QIDS).8

Treatment for complicated grief

The morbidity from CG can be long-standing, even for decades during which those so afflicted often describe either multiple failed treatments or falling into a chronic pattern of avoidant behavior or preoccupation with thoughts and behaviors related to their lost relationship with disbelief, anger, bitterness, intense yearning, or frequent reveries imagining their lost relationship that excludes outside influences that might challenge their assumptions or nudge them in other, more restorative directions.

Shear and colleagues developed a targeted treatment for CG called complicated grief treatment (CGT) that borrows from interpersonal psychotherapy (IPT)10 motivational interviewing,11 as well as cognitive behavioral therapy (CBT)12 to assist victims with the traumatic aspects of their loss that resemble PTSD. The techniques of CGT were tested and refined in a pilot study resulting in a 16-to 20-visit paradigm that was then applied in a randomized controlled trial comparing CGT and IPT.2 Inclusion criteria were: 6 months or more from their loss, and an ICG score of 30 or greater. If subjects were taking antidepressants at the time they were being evaluated for study participation, they needed to be stable on the antidepressant medication for at least 3 months, with at least 6 weeks on the same dose that was then continued unchanged for the duration of study participation. CGT was hypothesized to reduce symptoms of CG as measured by the ICG more completely and more quickly than IPT. Shear et al recruited an N of 49 randomized to receive CGT and 46 assigned to IPT with no baseline differences between groups in age (mean age 49), gender, race, education, marital status, type of loss, baseline ICG score, depression severity (measured by the Hamilton Depression Rating Scale13 and the Beck Depression Inventory or BDI,14 years since the loss (mean approximately 2 years), or the proportion who met criteria for comorbid major depression (approximately 20%), comorbid PTSD (approximately 25%) or the proportion whose loss was due to a violent death (approximately 15%). Results showed that, in the completer analysis, CGT was statistically superior to IPT in reducing ICG scores (from a mean of 46.4 to 25.8) as well as in reducing depressive symptoms on the Beck Depression Inventory (from a mean of 24.6 to 11.9) and the Work and Social Adjustment Scale15 (from a mean of 21.5 to 11.4).

The subjects over age 60 in this cohort were more likely than their younger counterparts to complete treatment, and the reduction in ICG scores measuring grief intensity in the CGT group was about twice that of the IPT group (personal communication with Kathy Shear).

With these encouraging results, a grant was obtained for a multisite RCT (Pittsburgh, New York City, Boston, and San Diego) named the HEAL study (Healing Emotions After Loss) with a goal of recruiting 350 subjects aged 18 to 95 with complicated grief for double randomized assignment to CGT vs TAU with an empathic psychiatrist who also administers randomly assigned citalopram vs placebo under double-blind conditions. This randomized controlled trial is underway at this printing, and thus no preliminary results are available at this time.

Sophia agreed to be a pilot subject in open treatment receiving CGT. She was not taking any psychotropic medications and was opposed to taking them.

Her CGT began by her trained CGT therapist listening to a detailed account of her husband's death. An indepth exploration of her relationship with her husband was sought as well as similar exploration of her other important relationships with parents, grandparents, siblings, and her own children. Sophia's reactions to prior deaths were also explored. By the third weekly visit, she was asked to bring in a significant other who was willing to hear all about the study and the nature of grief and to be supportive to the treatment process that Sophia was embarking on. Her brother attended, and was also asked to describe his view of the relationship between Sophia and her late husband and the changes he noticed in Sophia since his death. She chose to bring her brother as he knew both of them well and lived nearby. The steps in the treatment were also outlined for this special guest who would also hopefully serve as an empathic support during the treatment process, particularly in helping the griever to practice pleasurable rewards chosen by the griever as an incentive to endure confronting their own strong emotional arousal. Sophia was trained to rate her grief intensity on a 0-10 scale of intensity known as Subjective Units of Distress, or SUDS, and was asked to start keeping track of her SUDS with daily grief monitoring during the week between sessions. When her therapist reviewed these written self-reports, she paid special attention to any variability in SUDS scores during the week and clarified what appeared to be triggers for more intense grief and what actions tended to ameliorate her grief intensity, with the goal of giving Sophia a sense of better control of her grief and of inducing hope that it can be diminished in intensity with further effort in the therapy.

Subsequent sessions used specialized techniques such as imaginal revisiting in which Sophia was asked to close her eyes and describe what she saw and experienced right at the time of death of her husband for a brief but often intense period during which her therapist gave her encouragement and also asked her to call out her SUDS scores at various points in time over a 10- or 15-minute period. Her therapist then asked Sophia to imagine rewinding a videotape of the story and then to imagine putting the “tape” away in a safe place (this part of the exercise is intended to help convince Sophia that she can revisit strong emotions surrounding her husband's death and then put them aside without the fear of being completely overwhelmed). Her therapist then asks her to open her eyes and reflect upon being able to think about what she experienced during the imaginal revisiting session with emphasis on those elements that seem new or thoughts and circumstances that were keeping her stuck in her grief. The imaginal revisiting sessions were audio taped and Sophia was asked to listen to the entire tape once daily at a time of her choosing and to self-rate her SUDS level before, during, and afterward for each day of the week. This collected data is then reviewed and any trends downward over time in SUDS scores is pointed out as evidence that the intensity of her grief is being modified or improved upon. The imaginal revisiting exercise was repeated once weekly with new experiences coming to light each time; this reflects the work Sophia was doing internally to process all the emotions related to her husband's death that were not otherwise being accessed due to her avoidance. After 4 weeks of imaginal revisiting sessions, Sophia's SUDS scores dropped from 9s and 10s to 2s and 3s, indicating a freer acceptance of the facts of her husband's death.

Simultaneous efforts on personal goal work focused on new activities, the pursuit of lifelong dreams or fulfilling projects that would potentially give joy and satisfaction to her life, and emphasis was placed on taking concrete verifiable steps toward achieving them. Written reminders of these restorative efforts each week pinned Sophia down about what steps she agreed to carry out. She focused on doing more of the volunteer work that she loved but had fallen away from as well as beginning to discuss with her son which of her husband's clothes might be offered to others his size who could use them. These efforts were not easy for Sophia, and her therapist needed to encourage her to stay on task and to confront her when she did not complete her stepped task as she had agreed to the prior week. Other techniques such as imaginal conversation with the deceased, visiting the graveside, and preparing strategies for dealing with anniversaries and holidays were explored in depth using SUDS scores to “check in” with Sophia's level of internal grief intensity when these events approached.

Sophia made good progress over 16 weeks of CGT and her ICG score dropped to 23 and her QIDS score to 5. Sophia said the thing she missed the most was coming home to tell her husband all about her day's activities and, although she felt demoralized before CGT to the point of considering suicide despite having other caring family remaining, she said what helped her the most in CGT was learning to take stock of her life with its current limitations (including some of her own health problems), to become closer to her children and grandchildren, to volunteer more and to schedule theater trips and other entertainment with a group of other women, some of whom were also widows and who were “worse off than me.”

Complicated grief, major depression, and antidepressant medication

It has been said that “grief is not a disease but it can become one.” In DSM TV-TR, a diagnosis of major depression is excluded in the context of grieving within the first 2 months. Furthermore, common features of grief such as sadness, social withdrawal, sleep disturbance, rumination, and loss of pleasure in usual activities can also be symptoms of depression, which often perplexes primary care physicians about whether to treat a natural phenomenon with antidepressant medication versus not treating disabling psychopathology. Several studies have shown that antidepressant medication can relieve depressive symptoms that are bereavement-related. For example, Zisook treated bereaved individuals with major depression with bupropion and saw a robust response in reducing depression but only modest decreases in grief severity.16 Zygmont applied a combination of grief-focused psychotherapy and paroxetine in 15 subjects with complicated grief also found a modest reduction in grief intensity.17 Pasternak18 randomized depressed elders to nortriptyline vs placebo and showed that although depressive symptoms improved significantly better in the nortriptyline-treated group compared with those receiving placebo, the intensity of grief symptoms measured by the Texas Revised Inventory of Grief19 did not change appreciably. Reynolds et al randomized elders with bereavement-related major depression twice to receive nortriptyline vs placebo and also IPT vs Clinical Management (empathic support but no specific therapy) and showed that the nortriptyline-treated group showed a significantly higher remission rate than the placebo group but no added benefit for IPT (probably due to an undetectable effect size in this sample of only 80 subjects).1 In the cohort comparing IPT and CGT for CG mentioned earlier,2 Simon and colleagues analyzed the contributing effect of antidepressant medication, which was allowed to continue unchanged during the study if it was already prescribed at baseline. She found that those who remained on antidepressant medication were less likely to drop out of their treatment assignment to CGT (91% vs 58% completed the study) compared with assignment to the IPT group (70% vs 77% completed the study) suggesting that antidepressant medications, in addition to treating any concomitant major depressive symptoms, might also ameliorate the painful aspects of revisiting memories related to the death of their loved one that is an integral part of the CGT process.20 Further analysis by Simon et al showed that among those assigned to CGT, those concurrently on antidepressant medication showed a 61% response rate using a Clinical Global Improvement Scale (CGI)21 score of 1 (very much improved) or 2 (much improved ) vs 41% of those who were not receiving concomitant antidepressant medication.20


The utility of the concept of CG helps to differentiate those whose grief appears to be stuck, and whose suffering and debilitation is unremitting for extended periods of time, even decades. Agreeing on a final set of diagnostic criteria for complicated grief for inclusion in DSM-5 is the first step for broadening this distinction for wider application. With the aging of the babyboomer generation, there will be an increase in the proportion of grievers, a subset of whom will meet criteria for CG and thus require specialized treatment to be able to return to premorbid levels of function. CGT has shown promise toward the goal of restoration in one RCT comparing it with IPT, and we await the results of an ongoing larger RCT to further our knowledge regarding the interplay of CGT, antidepressant medication (citalopram), and their combination. The experience gained in this multisile study is also an opportunity to continue to refine the complicated grief therapy techniques and learn more about who will benefit most from them. Lastly, teaching paradigms will need to be developed for disseminating the finalized version of CGT as an effective treatment strategy for relieving the debilitating symptoms of CG.