SUMMARYMortality rates following patient admission and patient death were measured in the surviving spouses of a sample of dementia patients admitted to a psychogeriatric unit between 1971 and 1986. The vital status of 85% of spouses was confirmed, and the ratio of observed to expected deaths (SMR) was calculated using regional and national mortality statistics.Spouse mortality following first admission to hospital was significantly lower than expected (p < 0.006), and spouse deaths following bereavement were significantly fewer than expected (p < 0.026). When male and female spouse mortality rates were compared, no significant differences were found, but there was a trend towards lower mortality in female spouses.The results give no support to the theory that hospital admission often follows a breakdown in the health of a caring spouse. Previous studies have shown an increase in mortality following conjugal bereavement, but the spouses of dementia patients were relative survivors and outlived their married and widowed counterparts in the general population. The possible relationships between dementia, anticipatory grief, and the impact of bereavement on the health of the surviving spouse are discussed.KEY woms-Dementia, spouse, bereavement, mortality.The historical observation of high mortality rates in the widowed (Durkheim, 1951; OPCS, 1971) has stimulated research into mortality and morbidity following bereavement.
MORTALITY AFTER BEREAVEMENTMortality rates in the bereaved are higher than those in the married. Evidence of a 'mortality effect' during the initial 12 months following bereavement is stronger for widowers than for widows. Young et al.( 1 963) found a 40% increase in mortality during the initial six months of bereavement in a group of 4486 widowers aged 55 years and over. Parkes (Parkes et al., 1969) was able to explain the rise in mortality by an increase in cardiovascular deaths which was confined to the six months following bereavement. Rees and Lutkins (1967) confirmed Young's observations in a Welsh rural population and presented evidence of higher mortality rates following bereavement proportional to the nearness of the blood relationship. Helsing and Szklo (1981) were not able to show an increase in the female mortality rate following bereavement, although they did find evidence of significantly greater mortality after bereavement in widowed males. Clayton (1974) did not find evidence of higher mortality rates in a controlled 12-month follow-up study of widows and widowers, but the numbers in this study were insufficient for adequate postbereavement analysis of mortality rates.
MORBIDITY AFTER BEREAVEMENTMorbidity studies have tended to rely either on the uncertain relationship between consultation rates and true morbidity or on self-report questionnaires on health after bereavement. Parkes (1964) found