This study has an experimental design with two randomized groups, in which the effect of the intervention through a cognitive-behavioral group therapy (CBGT) is compared to a PEE in complicated grief, and symptoms of depression, anxiety, hopelessness and mental health in relatives of deceased cancer patients. Pre-treatment, post-treatment, and follow-up measures at 6 and 12 months after treatment were taken. The study took place between January 2018 and March 2020 at the La Paz University Hospital in Madrid. The project was approved by the ethical research committee of the La Paz University Hospital in Madrid.
The sample consisted of relatives of deceased cancer patients who had symptoms of grief and were referred to the Oncology and Palliative Care Unit of the Hospital Universitario La Paz by the Palliative Care and Primary Care units in Area 5 of Madrid. All participants carried out a clinical interview to confirm the presence of complicated grief symptoms. The inclusion criteria were the following: being over 18 years of age, speaking Spanish, having a family member who has died of cancer 6 months or more prior, availability to attend all assessment and/or treatment sessions, and scoring 25 or more in the Inventory of Complicated Grief. The exclusion criteria were: presenting psychopathological disorders or being under psychiatric treatment, substance addiction (according to the medical history), or undergoing any other grief treatments.
Two hundred and forty-nine relatives of deceased cancer patients met the inclusion criteria. A 48.7% were referred by the Palliative Care Units and 35.8% were referred by Primary Care in Area 5 of Madrid. Eighteen individuals in the experimental group did not complete the treatment and 35 individuals in the control group did not attend all the sessions and 8 did not fill in the questionnaires adequately. Therefore, the final sample consisted of 187 grieving relatives, 106 in the experimental group and 81 in the control condition. The statistical power with this sample size is 0.99 (calculated by the Gpower program with an effect size of 0.25 and alpha of 0.05)
After the selection and random assignment of the participants, groups of 8 to 10 relatives homogeneous in terms of age and type of loss were formed. Both treatments, CBGT and PSDEEI, were carried out, in both groups, by two clinical psychologists and were carried out across 16 weekly sessions, of 2 hours each, during a period of 4 months. All participants were assessed at baseline, one week after the end of treatment (post-treatment) and after 6 and 12 months of follow-up. The research was carried out in a period of 18 months, running three weekly intervention groups.
The treatment program is based on treatment programs that have been shown to be effective [14, 15]. The components of the treatment are the following: a) psychoeducation on grief and complicated grief; b) identifying and addressing cognitive, affective, physiological and behavioral manifestations, especially guilt; c) facilitating or favoring the acceptance of the reality of the loss, sharing personal experiences, reviewing the history of death, its implications and consequences, and searching for meaning/purpose; d) identifying and regulating emotion; e) improving self-care; f) facilitating optimal interpersonal functioning and communication; g) working on personal values and goals; h) Reviewing places and activities that are avoided; i) working with memories, images; j) imaginary conversations and letters to the deceased. Relaxation techniques, breathing, in vivo exposure and imagination, cognitive modification and behavioral activation were used.
Groups of 8 to 10 people were formed following the same criteria as the experimental group. They received a total of 16 weekly sessions of 2 hours duration, just like the experimental group. The content of the sessions consisted of psychoeducation on grief and emotional ventilation.
Variables and Instruments
A semi-structured interview was carried out to collect sociodemographic data, referral data and clinical history, drug treatments, personal grief reactions, personal resources, and socio-family support.
The following instruments were used to measure the efficacy of the treatment:
Beck Depression Inventory (BDI-II) (Spanish version) . The cut-off point to consider high levels of depressive symptoms was 21 in the present study. A score between 0 and 13 would indicate a level of minimal depressive symptoms, between 14 and 19, mild, between 20 and 28, moderate, and between 29 and 63, severe.
Beck Anxiety Inventory (BAI) (Spanish version) . The cut-off point to consider high levels of anxiety symptoms was 21 in the present study. A score of 0-7 would indicate minimal anxiety, 8-15, mild anxiety, 16-25, moderate anxiety, and 26-63, severe anxiety.
Beck Hopelessness Scale (BHS), Spanish adaptation . The severity categories are as follows: 0-3- Mild hopelessness within the limit, 4-8 slight hopelessness, 9-14 moderate hopelessness, 15-20 severe hopelessness. In the present study, >= 8 was used as a cut-off point.
Goldberg's General Health Questionnaire (GHQ28) [19,20]. It is a screening instrument that aims to detect psychological morbidity and possible cases of psychiatric disorders in contexts such as primary care or in the general population. It is a mental health assessment measure. Scores > = 6 were considered indicative of psychosocial problems.
The Inventory of Complicated Grief (ICG – Inventario de Duelo Complicado, IDC in spanish) [21,22]. The cut-off point to consider complicated grief was => 25 at six months after death.
The distribution of the qualitative variables was presented in the form of absolute and relative frequencies and the distribution of the quantitative variables was described by the mean and standard deviation or median and interquartile range, depending on the distribution of the data.
A univariate analysis was carried out to study the homogeneity of the two groups in terms of their sociodemographic characteristics and the baseline situation of the scales: to contrast the differences of the qualitative variables, the chi-square test was performed and to compare the differences between the quantitative variables, the Student's t test for two independent samples or the nonparametric Mann Whitney U test, depending on the distribution of the data.
To study the efficacy of the treatment, the development of the BDI, BAI, BHS, GHQ28 and ICG scales over time was analyzed by adjusting Linear Mixed Models including the treatment (Experimental vs Control) as a fixed factor and the measurement time as a repeated measures factor (Pretreatment, Posttreatment, 6 months, and 12 months). The Treatment x Time interaction effect was also included, with a variance matrix without structure. A statistically significant interaction effect indicated that the groups differed in their development. In the hypothesis tests for post-hoc comparisons, the Bonferroni correction for multiple comparisons was used. The eta-squared value for the interaction effect was also presented as a standardized measure of effect size. This statistic describes the proportion of total variability attributable to a factor.
All tests were considered two-tailed and a p-value less than 0.05 was considered statistically significant.