Benchmarks for afterlife research
Dr Dewi Rees’s ‘Hallucinatory reactions of bereavement’, published in the British Medical Journal in 1971, and the cross-cultural study on ‘Deathbed observations by physicians and nurses’ undertaken by Karlis Osis and Erlunder Haraldsson remain benchmarks for evaluating contemporary understandings of near to death experiences and ongoing research, although both these studies were undertaken over thirty years ago.
The large-scale ‘Deathbed observations by physicians and nurses: a cross-cultural survey’ undertaken by Karlis Osis and Erlendur Haraldsson, was published as At the hour of death (1977). The survey was conducted in some north-eastern states of the United States between 1961 and 1964, a second part was conducted in northern India during 1972−73, both based on an earlier pilot survey undertaken by Karlis Osis, which, in turn had been inspired by the work of Sir William Barrett on deathbed visions.
The original pilot survey questioned physicians and nurses about the experiences of dying people, particularly concerning hallucinations or visions and found that people who were dying did indeed have hallucinations. While some had a pathological origin, in general this was not the case. This survey found that these hallucinations were different to those caused by illness, mental or physical, in that they were shorter, more coherent and were related to the experience of dying. In this survey, people who were dying felt that the figures seen in the hallucinations had a particular purpose—to ease the transition from this world and to ‘take them away’. Studies of hallucinations in the general population have found that those who experience hallucinations more commonly experience visions of living rather than those who are deceased—this survey found the reverse in the case of people who are dying with 83% experiencing visions of deceased close relatives.
In the later 1961 and 1964 US and the 1972−3 Indian survey Osis and Haraldsson tested for demographic, religious affiliation and psychological factors, such as fear of death, desires and expectations. They concluded that factors such as age, sex, education and occupation did not interact significantly. Religion did not seem to influence significantly the purpose or the kind of hallucination seen (living, dead or religious figures), with the occurrences of visions (of the dead/religious figures) and their afterlife purpose appearing to transcend the widely divergent religious ideologies of Hindus, Catholics, Protestant, Jews and Muslims.
There were, however, cultural differences. The identity of the apparitions varied greatly between the groups in India and the USA. American patients mainly saw deceased persons, while Indian patients predominantly saw religious figures. Osis and Haraldsson assume that individual and cultural factors will completely shape deathbed visions if they are caused by these factors. If, however, these experiences were based on some external reality, then only minor cultural differences would emerge. This research concludes that, while cultural ‘colouring’ is present, the main findings of the three surveys are consistent and support the afterlife hypothesis proposed in the pilot survey.
A more recent study on deathbed visions in India was conducted by palliative care physicians (Sandhya Muthumana and colleagues) and published last year in Omega Journal. While the Osis and Haraldsson study collected reports from doctors and nurses, this research was based on family observations of the last weeks and days of a family member who was dying. In all, 104 families were interviewed and of these forty families reported ‘unusual experiences and behaviours’ from the person who was dying—thirty were consistent with deathbed visions such as interacting or speaking with deceased relatives. There were also six cases of reported premonitions of death. As with the Osis and Haraldsson study, factors such as gender, age and occupation were not found to be significant.
There was however, an interesting difference between these two Indian studies. Osis and Hraldsson had reported the appearance of Yamdoots—messengers or supernatural figures given the task of collecting the souls of the dead to take to the other world in Hindu religious mythology. Reports of Yamdoots were present in 18% of the Osis and Haroldsson Indian reports but made no appearance in this more recent study. The researchers suggest some reasons for this—possibly the geographical difference as the Osis and Haraldsson study was conducted in northern India and the latest study was conducted in Kerela in the south. The literacy rate in Kerala is the highest in India—at around 90%—this educational level may moderate traditional beliefs in Yamdoots. Beliefs may also have declined in the intervening years between the two studies.
There is also the question of who is reporting these experiences and inevitably there can be a degree of reporter bias. For example, as Muthuman suggests, in the Osis and Haraldsson study reports were collected from doctors and nurses some of whom may have offered more traditional descriptions or interpretations of their observations, especially to western researchers. Muthumana also acknowledges potential bias in under reporting the presence of Yamdoots as their data was collected by Indian clinicians and researchers and local families may not have disclosed their traditional beliefs to high status health care professionals in their locality.
The presence of cultural ‘colouring’, within and between similar and very differing cultures, is an important consideration when trying to understand near to death experiences and the meaning they may hold for people. Methodological differences also play their part; the Muthumana study estimates the prevalence of these experiences at around 30% amongst people who are dying; Osis and Haroldsson estimated 25% (from reports amongst doctors and nurses) and Fountain (published in Palliative Medicine in 2001) reported 47% of visual hallucinations amongst terminally ill patients
Notwithstanding these differences and perspectives there are clearly large numbers of dying people experiencing a range of near to death experiences and with family members and health care professionals involved in their care, even greater numbers of people becoming aware of these experiences.
All the more reason to be grateful for the benchmark provided by Osis and Haroldsson, which guides the way forward for future research and the potential to harness these experiences to support people who are dying and these who care for them.