Table 2.
Results Table.
| Author (year) Country | Study design | Number | Participant demographics | Summary of results | Study limitations |
|---|---|---|---|---|---|
| Bolton & Camp (1987) U.S. | Study design: observational cross-sectional quantitative surveyPopulation: widowed persons selected at random from previous clients of Widowed and Family Grief Counselling ProgramIntervention: self-reported number of pre-, during-, and post-funeral ritualsOutcome measures: two measures of grief adjustment – Affect-Balance Scale and Attitude Inventory | 50 | Mean age 55.6 years94% femaleAverage of 6 years since death of spouse | “No statistically significant degree of association for the major variables were produced” i.e. no statistically significant relationship between number of pre-, during-, and post-funeral rituals and grief adjustment measures. No reporting of interpretive statistical data to support this “Subscales of the Attitude Inventory (usefulness, health, happiness, and financial) were clearly related to post-funeral rituals such as sorting personal effects, removing the wedding ring, visits to the grave side, and the disposal of personal effects”. No reporting of interpretive statistical data to support this | Random selection process not specified, small sample size, almost all female sample, time since bereavement variable, interpretive statistical data on primary outcome and subanalysis not reported |
| Doka (1985) U.S. | Study design: observational cross-sectional mixed methods – quantitative survey with qualitative interviewPopulation: survivors who had primary responsibility for arranging funeral rituals of a death that occurred 12–18 months prior to the onset of the study obtained from referrals from students, clergy, senior citizens groups and funeral homesIntervention: self-reported participation in planning and conducting funeral ritualsOutcome measures: adjustment to death using a modification of the Carey Adjustment Scale | 50 | 100% white middle to upper class76% femaleAge: 12% 18–35 years, 45% 36–60 years, 28% ≥61 years60% Protestant, 32% Catholic, 8% Jewish | “There were no significant differences between involvement [in planning and conducting of funeral rituals] and grief adjustment a year later” Chi-Square value for planning of funeral rituals and grief adjustment: 1.09, p-value not reportedChi-Square value for participation in conducting funeral rituals and grief adjustment: 0.94, p-value not reportedCatholics were significantly less likely to report participation in planning funeral rituals that Jewish or Protestant respondents (Chi-square value: 6.75, p < 0.01) Participating in planning funeral arrangements: 57% of those who participated in planning felt it helped them with their grief, while another 28% were unsure. Some felt the busyness itself was worthwhile. Others found that the intensive involvement with caregivers was comforting. Others derived comfort from personalizing and being part of the service. 2 participants found that involvement reaffirmed their own abilities and worth. The 15 percent that did not define the planning process as helpful were not negative toward participation but tended to look on the task as a routine duty associated with funerals. Half of the 20% who did not assist in planning defined themselves as emotionally incapable of doing so at the time. Only 20% of the sample reported actual participation by themselves or other family members in the funeral service itself. All but one respondent reported that this too eased grief. | Small sample size, ethnically, socially and religiously homogenous sample, predominantly female sample |
| Fristad et al. (2001) U.S. | Study design: observational prospective cohort quantitative surveyPopulation: parent-bereaved children aged 5–17 years recruited from obituaries from local newspapers and contact with local funeral homes for the Grief Research StudyIntervention: participation in the visitation, funeral and burial using The Funeral Questionnaire – Child and Parent FormsOutcome measures: grief measured using The Grief Interview – Child and Parent Forms, depressive symptomology measured using The Child’s Depression Rating Scale – Revised and Diagnostic Interview for Depression in Children and Adolescents, symptoms of 16 psychiatric disorders using The Diagnostic Interview for Children and Adolescents, and overall psychiatric symptomology using BAMO scale. Outcomes measured at 1, 6, 13- and 25-months post-death | 318 | 59% age 5–12 years, 41% age 13–17 years98% Caucasian | “Nearly all children whose families had visitations, funerals, and burials attended. Thus, comparisons could not be made between those children who attended and those children whose families had the ritual but did not attend … Therefore comparisons were made between 258 children who attended a visitation and the 38 children whose families did not have a visitation” No differences were found between groups at 1 and 6 months post-parental deathBy 13 months post-loss, overall symptomology was 50% lower for children who did versus did not attend the visitation ( (0.6 ± 0.8 vs. 1.2 ± 1.5; t 2.26, df 37.7, p <.03) as well as depressive symptom severity (22.3 ± 7.7 vs. 30.6 ± 13.4; t 2.36, df 15.4, p < .03) By 25 months post-loss, those children who had attended the visitation had fewer PTSD symptoms than those who did not (0.4 ± 0.7 vs. 0.7 ± 0.8; t 2.08, df 188, p < .05). | Ethnically homogenous sample, analysis changed to suit participants not pre-specified, clinical significance not considered |
| Gamino et al. (2000) U.S. | Study design: observational cross-sectional mixed methods – quantitative survey with qualitative interviewPopulation: bereaved individuals participating in the Scott & White grief study: recruited from outpatient psychiatry clinic; family and friends sent a condolence letter after loved one died in hospital; self-help/grief support groups; personal contact from investigatorsIntervention: self-reported participation in funeral planning, funeral/burial service attendance, perceptions of funeral experience, whether funeral service was described as comforting, any adverse events reported in connection with funeral/burial ritesOutcome measures: grief symptomology measured by Grief Experience Inventory | 74 | Mean age 50.7 years78.4% female91.9% white, 4.1%African, 4.1% Hispanic4.1% no religion, 28.4% mainline Protestant, 44.6% conservative protestant, 1.4% Pentecostal, 18.9% Catholic, 2.7% Jewish | Mourners who described funeral/burial services as “comforting” reported significantly less overall grief (F = 5.33, p = .01) and subscales of social isolation (F = 7.28, p = 0.005), despair (F = 5.34, p = 0.01), anger/hostility (F = 4.04, p = 0.02) and guilt (F = 2.93, p = 0.05) “Nearly every death was followed by a funeral or memorial service … Among those mourners with the opportunity to attend services, almost all chose to do so. Therefore, no meaningful statistical distinction could be drawn between [those who attended and those who did not]”. Those who participated in planning the funeral reported significantly lower depersonalisation (F = 4.10, p = 0.001) and social isolation (F = 2.91, p = 0.05) than those who did notThose who experienced adverse events (e.g. conflicts among survivors, discrepancies between the expressed wishes of the decedentand the preferences of the survivors, issues with cremation, state of the body, problems with the funeral home, problems with the minister, financial problems) during the funeral service had significantly higher overall grief (F = 3.45, p = 0.05), and subscales of somatization (F = 10.73, p = 0.001), loss of control (F = 4.84, p = 0.02) and depersonalization (F = 2.89, p = 0.05) | Ethnically homogenous sample, predominantly female sample, time since bereavement variable, clinical significance not considered |
| Grabowski & Frantz (1993) U.S. | Study design: observational cross-sectional quantitative surveyPopulation: volunteer participants of Latino and Anglo origin who had experienced the death of a relative, friend or acquaintance Intervention: funeral attendance or novena participation (nine-day post-funeral practice involving prayer and support that generally takes place in the home of the bereaved) Outcome measures: intensity of grief reactions measured by Texas Revised Inventory of Grief (TRIG) – Part I measures adjustment to past life event i.e. death, Part II measures intensity of present feelings of grief over loss of a loved one | 50 Latino and 50 Anglo participants | Of all 100 participants, 95% Roman Catholic69% femaleMean age 47 years | No significant difference in grief intensity between those who did and did not attend the funeral in Latino and Anglo samples (F = 0.5, p value not reported) In the Latino sample there was no significant difference in grief intensity between those who had and had not participated in a novena (F = 1.11, p value not reported) In the Latino sample who had attended a novena, there was no significant correlation between their self-report of helpfulness of a novena and grief intensity (t = 01.506 for Part I and t = −0.932 for Part II of TRIG) | Religiously homogenous sample, volunteer sample, TRIG in English only, time since bereavement variable |
| Kissane, et al. (1997) Australia | Study design: observational prospective cohort quantitative surveyPopulation: bereaved spouses with one or more children aged 12 years or older of a relative who died from cancer when aged 40-65 yearsIntervention: use of mourning rituals and “relevant aspects of the death and funeral [including] saying goodbye and viewing the corpse” using an interview with the bereaved spouse including qualitative and quantitative components on four-point Likert scalesOutcome measures: grief measured by Bereavement Phenomenology Questionnaire (BPQ), psychological morbidity using cognitive items of Beck Depression Inventory (BDI) and Brief Symptom Inventory (BSI), social functioning using Social Adjustment Scale (SAS), at 6 weeks (T1), 6 months (T2) and 13 months (T3) following the death | 115 at T1, 104 at T2, 100 at T3 | Mean age 55.9 years53% female66% Australian, 11% English, 7% Eastern European, 5% Italian, 4% Irish, 2% Asian, 1% Greek, 4% other85% Christian, 3.5% Jewish, 8% no religion | Not viewing the body of the deceased correlated with BDI i.e. more depressive symptoms at T1 (Pearson correlation 0.3297, p < 0.001) and negatively correlated with BPQ i.e. more grief intensity T1 (Pearson correlation −0.3905, p < 0.01). However due to small numbers this variable was not included in best subset regression analyses. Saying goodbye as wished correlated with SAS i.e. better social adjustment at T1(Pearson correlation 0.2634, p < 0.01). This variable was not included in best subset regression analysis, reason unspecified. Neither of the above variables correlated with any psychological outcomes at T2&3. “Experience of the funeral and mourning rituals … failed to influence bereavement outcome” – no data provided to support this | Only including nuclear families, religiously homogenous sample, dropout characteristics not identified, unclear reporting of outcomes, qualitative data collected not reported |
| Mitima-Verloop et al. (2019) Netherlands | Study design: observational prospective cohort quantitative surveyPopulation: individuals bereaved within the last 6 months recruited via routinely administered customer satisfaction survey of a funeral service company Intervention: perception of the funeral using Funeral Evaluation Questionnaire (FEQ), grief rituals using derivative of Bereavement Activities Questionnaire (BAQ) Outcome measures: grief using Traumatic Grief Inventory self-report version (TGI-SR), positive and negative feelings using Positive and Negative Affect Scale (PANAS), impairment in functioning using Work and Social Adjustment Scale (WSAS) at T1 (time of first survey) and T2 (3 years later) | 552 at T1, 289 at T2 | At T1: Mean age 58.9 years58.5% femaleNationality and religion collected at T2: 97.6% Dutch (without migration background), 2.4% other29.6% Christian, 16.7% Spiritual, 50.2% no religion, 3.5% other | Participants perceived the funeral as contributing to processing their loss (agreed with the statement “The way in which the period around the funeral was organized, was important in processing the loss” “a lot” to “very much” 75.9% at T1 and 70.2% at T2) with a high mean item score (M=4.07, SD=1.07 at T1 and M=3.92, SD=1.11 at T2).Positive association between general evaluation of funeral and positive affect at T1 (r-0.21, p < 0.001) and funeral director evaluation and positive affect (r=0.13, p = 0.003)Hierarchical regression analysis with grief and general evaluation of funeral and funeral director at T1 scores predicting grief scores at T2 was significant (F = 248.82, p < 0.001). However grief at T1 explained a unique proportion in variance in grief at T2 (β=0.696, p < 0.001) but not the other two variables (p = 0.596 and p = 0.283 respectively) | Recruitment from satisfaction survey, culturally homogenous sample, T1 to T2 dropouts significantly demographically different to T2 participants, FEQ designed for this study and not validated |
| Saler & Skolnick (1992) U.S. | Study design: observational cross-sectional quantitative surveyPopulation: adults who had experienced the death of one parent before the age of 18 years, and were now aged between 20 and 50 years recruited from public noticesIntervention: children’s participation in various mourning activities using The Mourning Behaviour Checklist (MBC) Outcome measures: parental attitudes and behaviour using Parental Bonding Instrument (PBI), depressive symptomology using The Center for Epidemiological Studies Depression Scale (CES-D), depressive experience using The Depressive Experiences Questionnaire (DEQ) | 90 | Mean age 32.2 years58% female43% Jewish, 26% Catholic, 14% Protestant, 7% no religion92% white, 8% Asian/black/Hispanic/other | MBC was the only statistically significant variable in multiple regression analyses determining contribution to CES-D (β=0.2876, p = 0.0133 when using PBI raw score and β=0.3212, p = 0.0067 when using PBI parenting style scores), MBC was significantly associated with higher Self-Criticism scores on DEQ (β=0.23, p≤0.05) i.e. those who reported less opportunity for participation in mourning activities had higher rates of depressive symptomology and were more prone to self-criticism | Ethnically homogenous sample, retrospective self-reporting of mourning activities, breakdown of specific question contributions not reported |
| Schaal et al. (2010) Rwanda | Study design: observational cross-sectional quantitative surveyPopulation: widows (who had not remarried) and orphans (lost one or more parents) over 18 years old who had experienced the Rwanda genocide in 1994Intervention: self-reported funeral attendanceOutcome measures: Prolonged Grief Disorder diagnostic status and symptom severity using PG-13, Post-traumatic stress disorder symptoms using the PTSD Symptom Scale Interview (PSS-I) | 400 | 87.7% femaleMean age 37.18 years61% Catholic, 23.3% Protestant, 4% Islamic, 2% Adventist, 6% other, 3.8% no religion | Multiple regression analysis with grief score as dependent variable showed funeral attendance did not significantly contribute to the severity of prolonged grief reactions (BPGD-score -1.14, B SEPGD-score 0.68, bPGD-score -0.06. p-values not given). | Losses due to violence may not be generalisable, predominantly female sample |
| Weller et al. (1988) U.S. | Study design: observational cross-sectional quantitative surveyPopulation: bereaved children agreed 6-12 years with a IQ ≥70 and no chronic incapacitating medical or psychiatric illnesses present and their surviving parent, and the following applied: at least one parent had been employed the majority of the time in the 2 years preceding death; no chronic incapacitating illness in either parent had been present in the 2 years preceding the death (other than that associated with the deceased parent's death); no family member had received inpatient or outpatient psychiatric treatment in those previous 2 years; the surviving parent was able to complete the questionnaires and be interviewed; children of divorcedparents had to have had frequent visitation with both parents; parental death was not caused by suicide or homicide. Local obituary section used, funeral home director/clergy contacted to discuss appropriateness of contacting family, if appropriate family contactedIntervention: children’s participation in funeral activities using Death Related Behaviour Questionnaire – Child/Adult FormOutcome measures: presence/absence of psychiatric diagnosis using The Diagnostic Interview for Children and Adolescents (DICA-C/DICA-P), grief experiences using The Grief Interview – Child/Parent Form | 38 children, 26 parents | Children: 47% maleParents: 73% female87% white, 8%H Hispanic, 5% black | “T tests were used to determine whether children’s participation in … funeral activities was associated with [depression or anxiety] symptomology … The two groups did not differ significantly in depressive, anxiety or other psychiatric symptomology as rated by the child or parent.” No interpretive statistics provided to support this | Very strict inclusion criteria, ethnically homogenous sample, predominantly female sample, interpretive statistics not presented |
| Zisook & DeVaul (1983) U.S. | Study design: observational cross-sectional quantitative surveyPopulation: friends and colleagues of the authors who had lost a relative or close friendIntervention: self-reported funeral attendanceOutcome measures: unresolved grief using Unresolved Grief Index | 211 | 62% female65% white, 17% black, 11% Mexican American, 7% otherMean age 36.5 years47% Protestant, 26%Catholic, 13% Jewish, 13% other/none | Participants with “definitely unresolved grief” (score of ≥6 on Unresolved Grief Scale) were less likely to have attended the funeral (p < 0.05) | Recruitment method, unvalidated questions measuring outcome, interpretive statistics not specified |
| Aksoz-Efe et al. (2018) Turkey |
Study design: qualitative phenomenological semi-structured interviews Population: Turkish women who had experienced a death loss and early traumatic experience recruited by a college’s psychiatry department and psychological counselling guidance programme Aim: examining the definition and meaning of grief, examine possible participant perceived connections between cultural bound death-related rituals and beliefs and their grief |
8 | Age 25–59 yearsAll Muslim | Metaphors of lossDestruction in the lives of the bereaved, pattern and wholeness of the family corrupted, irreversibility and inevitability of deathFuneral ritualsFuneral rituals most commonly involved visits to the deceased’s house (the funeral house), which continue from 7–40 days after death to meet the physical, psychological and social needs of the bereaved. Specific rituals include praying from the Quran for 7 days, not turning on TV for 7 days, the 7th, 40th and 52nd days (i.e., visits for prayers), Mevlid (i.e., spiritual poems about the prophet Muhammed), turning on the light of the room in which the person died, and taking some souvenirs from the deceased. Importance of seeing the body of the deceased. If the deceased’s body has to wait in the home, it is covered with a white sheet called kefen which can be opened just to leave the face and sometimes the hand out for the bereaved to see the deceased for the last timeRituals in relation to control, age, socioeconomic status and griefRituals either deemed helpful or unhelpful depending on participants’ perceived control over rituals: when religious practice based on own will, experienced relief; when imposed, felt guilty and distressed. Younger, higher socioeconomic status women experienced more control as they were employed and therefore had the ability to be physically away from home. The specific elements of ritual that emerged as most critically related to participants’ grief were religious beliefs, condolence visits, talking about the deceased, and cultural expectations of when grief “should” end. Praying, thinking that their loved ones were good people and believing that they will be in Heaven were helpful. Participants also reported unhelpful aspects related to religious beliefs andpractices. In some cases, when participants were questioning their beliefs after the loss, others were perceived as putting pressure on them about not losing their faith. Because of the seemingly imposed behaviours of others, some of the participants described feeling guilty and distressed for not fulfilling their religious duties. The imposed religious expectations were viewed as new and additional burdens by the participants. After a while, usually the first week, participants described feeling stressed about visits and crowds, and need to be alone and have their own time to focus on feelings and thoughts. Participants were unable to experience their grief due to “preaching” messages of others. Concern about judgement among those with no personal space or control over their grief. All participants reported being particularly concerned and stressed when other people were around. After 30–40 days, people expected them to start acting normally. Participants described being encouraged to stop grieving, continue with their lives and focus on daily life responsibilities. Again, helpful or unhelpful depending on sense of control | Variable time after loss |
| Chan et al. (2005) Hong Kong | Study design: retrospective qualitative analysis of individual bereavement counselling interviewsPopulation: bereaved former clients of a community-based bereavement centre aged over 18 years who received counselling within the last 2 years, not recruited for this study, former counselling recordings analysed Aim: explore the bereavement process of Chinese persons in Hong Kong and examine the influence of the Chinese culture on the experience of the bereaved | 10 counselling sessions analysed; themes then compared to 42 other transcripts | 78.85% female44.23% no religion, 25% traditional ancestor worship, 9.62% Buddhism, 21.15% Christianity | Meaning makingCause of death categorised in terms of traditional beliefs (fate, karma, fate clashes among family members, feng shui, evil spirit). Presence of family at moment of death appraised positively and negatively by equal numbers. In Chinese culture, there is a strong belief in the afterlife. Most had a positive sense of the destination of the deceased in the afterlife. Participants mentioned that their relationship with the deceased (yuan) would be continued. Two thirds of participants perceived that it was good to participate in the funeral to have the final change to say goodbye to the deceased. The remaining third harboured negative feelings, largely because they had not been able to participate. During a Taoist funeral ritual, the lay priest (Nahm Mouh) chants and performs sacrificial litanies. Similar rituals by Buddhist nuns and monks. Viewed positively mostly, expect when costly or contrasting with bereaved person’s belief system. Ritual viewing of body before final covering of coffin had some negative memories. “Soul-flag” and “purifying wash” (symbolically washing a piece of the deceased’s clothing) viewed positively by half, but negatively by half as they did not have a son who would usually carry out ritualsBond continuationContinuing the bond with the deceased anchored in traditional Chinese beliefs. Continuation of relationship initiated by either the deceased (feeling, hearing or seeing the deceased) or by bereaved (talking to a photo of the deceased, visiting graveyard or cemetery) | Therapeutic interviews used so not aiming to answer research question directly |
| Nesteruk (2018) U.S. | Study design: retrospective qualitative analysis of semi-structured interviews from a larger studyPopulation: immigrants in the United States aged 45 and older who arrived as young adults (aged 18-30 years) recruited through senior centres, residential facilities, personal contacts of the author and snowball samplingAim: advance understanding of long-term immigrants’ coping with death and bereavement in the transnational context | 56 | Mean age 6476.8% femaleOriginated from 31 different countries | Caregiving in transnational familiesParticipants described a sense of duty to either provide care directly or financially contribute to care from a distance. Sense of obligation and guilt when reflecting on aging loved ones in countries of origin. Participants described feelings of distress at their inability to attend a loved one’s funeral. Coping with loss and transnational grievingParticipants who were able to attend funerals in their countries of origin felt that being with their loved ones and participating in rituals associated with death, such as making funeral preparations, sharing meals, praying, and attending services, provided them with a sense of belonging and comfort. Grieving alone in adoptive country was challenging due to the lack of social support. Being surrounded by familiar rituals and being able to share one’s emotions with others who are also grieving helps one process the loss, not feel alone, and facilitate coping. Other participants emphasized that being remote from the death of a loved one provided an emotional barrier from the familial upheaval and cultural drama that is often associated with the end-of-life rituals in their countries of originIncreasingly, technological advances in communication enable immigrants to go online to witness the services of their loved ones, see family, and virtually participate in the cultural rituals associated with death. Although these technological advancements may not be commonplace in every situation, they do represent an opportunity for immigrants to virtually cross large distances and borders, and to be present at the funeral services for a loved one.Family discontinuity and anticipatory grief as mitigating factorsImmigrants lose physical contact, support, and the presence of family members which in time may lead to a discontinuity of family relationships. Over time this leads to a greater self-reliance and the ability to manage change and loss without family support. Loss may be mitigated by years of separation. Family of procreation and work duties demanded attention and distracted from loss. | Part of a large study not aimed at examining bereavement experiences in detail therefore not aiming to directly answer research question |
| Pang and Lam (2002) Hong Kong | Study design: qualitative semi-structured interviews Population: widowers who had taken part in either Chinese or Christian death ritualsAim: report widowers’ bereavement experiences and how the widowers go through the bereavement process by the performance of Chinese and Christian death rituals | 4 | Age 37–50 years2 Christian, 2 no religion | Funeral ritualsPersonalised elements of funeral rituals provided meaning and eased grief, provided a chance of catharsis to restore emotional stability. Benefited from console given by the funeral participants whose attendance confirmed the value of the decease. However, low attendance at funeral was interpreted as a lack of social support which contributed to feeling worse. Postfuneral ritualsWith the performance of post funeral death rituals, beliefs were reinforced and a continuing bond with the deceased was established | Recruitment process not specified, reporting of individual cases as opposed to working out commonality and themes |
| Silverman (1987) U.S. | Study design: qualitative interviewsPopulation: college-age women who had lost a parent to death recruited from a nearby graduate professional schoolAim: To examine female children’s reactions to the death of a parent | 18 | 11.1% blackMost respondents from upper middle-class families | Coping with deathWith one exception, all their families had traditional funerals. For the older women, seeing the body seemed to be important. Participant who did not felt it may have made grieving easier. Some women felt that their participation in the funeral made a difference and that the funeral brought them together as a family. Other women, especially several who were teenagers at the time, felt resentful of the way they were included and, in the long run, seemed to have more difficulty coping. Those who were younger at the time of death seemed reassured by being included among the mourners and knowing what to expect. For those that were already teenagers, they wanted to be involved in a more adult fashion in the decision making, but not in making the actual decisions. Although data support the idea that participation in the funeral and associated religious ritual is important, in the long run, parental responsiveness from the very outset was most critical, in the eyes of these women, to how they managed. Coping by trying to carry on as normal – by being busy they could avoid a confrontation with their new reality. However, difficulties with concentration and study. Other people’s concerns seemed trivial and frivolous. Increased isolation from peers, learning to keep their feelings to themselves. The bereavement processAll participants continue to deal with the death, no reporting of closure. As adults, they needed to renegotiate their relationships to their parents alive and dead. Acknowledgement of what they have missed and what the death meant to the surviving parent. | Recruitment process unclear, information about participants limited |
| Vandercreek & Mottram (2009) U.S. | Study design: qualitative semi-structured interviewsPopulation: suicide bereaved female individuals aged over 21 years who suffered the suicide of a family member at least 2 years in the past, recruited through suicide support groups (first 10 volunteers accepted) Aim: explore the function of survivor’s personal religion, support from family and friends, and established religious communities during bereavement | 10 | All Caucasian9 Christian, 1 Buddhist7 considered religion “very important”, 1 “somewhat important”, 2 “not important” | The functions of the survivor’s personal religionBereavement process led over time to a new, clearer religious purpose in life. The function of religious support from family and friendsMany families and friends expressed support through sympathy cards or verbally. Limited help as implied not willing to help wrestle with bereavement concerns. Meaningful support from those chosen by the bereaved or those who had also lost a child by suicide. The functions of established religious communitiesAll survivors held a funeral and many people attended. The survivors tended to interpret the attendance of these friends, acquaintances, and concerned strangers as religious support.A survivor found the expression of support at the funeral helpful even at the time of the interview many years after the suicide | Religiously homogenous sample |