Using the Mini-Mental State Examination to screen for delirium in elderly patients with hip fracture

BACKGROUND

The main aim of this paper was to examine the usefulness of the Mini-Mental State Examination (MMSE) for screening delirium in elderly patients with hip fracture.

METHODS

The sample included 364 elderly patients with hip fracture admitted to two hospitals in Oslo. Delirium was assessed by the Confusion Assessment Method (CAM) as an approximation of the gold standard, DSM-IV. To evaluate the psychometric properties of the MMSE scale, we used the Mokken nonparametric latent trait model for unidimensional scaling.

RESULTS

In total, 76 (21%) patients were diagnosed with delirium based on the CAM, and 141 (43%) had preexisting cognitive impairment. As a screening tool, the recommended MMSE cut-point of 24 showed an acceptable sensitivity of 88% with a specificity of 54%. The prediction of delirium, based on logistic regression on the MMSE total score and on the 5 items selected by the stepwise logistic regression procedure, gave clearly less acceptable results.

CONCLUSIONS

Our study indicates that the MMSE may be useful in screening for delirium, but the high percentage of false positives shows that it does not have diagnostic potential in patients with hip fracture.

Factors affecting grief reactions in close family members to individuals who have died of cancer

This longitudinal study examined factors related to grief reactions in a systematic and standardized way in 183 close family members to individuals who died of cancer. Grief reactions were measured using the Texas Revised Inventory of Grief (TRIG). A repeated measures MANOVA was used to test and compare the grief reactions of the bereaved for one year after the loss. The female respondents showed stronger grief reactions than the male respondents. The grief reactions increased with age, and those who had lost a younger family member experienced stronger grief reactions than those who had lost an older family member. The relationship to the deceased, the duration of the disease, place of death, aspects of social support such as children living at home, and employment were not related to the grief reactions in the bereaved respondents when controlling for the former factors.

The first year of grief and bereavement in close family members to individuals who have died of cancer

Using a systematic and standardized method this longitudinal study examines changes in grief reactions in a sample of close family members (n = 183) to individuals who had died of cancer. The respondents were followed for 1 year after the loss. The study sample originated from a cluster randomized trial evaluating comprehensive palliative care (intervention) against conventional care (control). Hence, we also compared grief reactions among close family members to the patients in the intervention and control groups. Overall, the family members' grief reactions, as measured by the second part of the Texas Revised Inventory of Grief (TRIG), showed a significant decline over the period studied. However, we found no significant differences in grief reactions between the family members to the intervention and control patients at any point in time, and the pattern of change did not differ significantly for the two groups.

A follow-up study of the quality of life in cancer patients with different prognoses

The present study examines differences in quality of life between two points in time in a heterogeneous sample of cancer patients with different prognoses. The original sample comprised 253 patients, while the follow-up data obtained 3-4 years later included 75 patients, that is, 92% of the survivors. All the patients completed a questionnaire including the EORTC QLQ-C30. The research questions focus on overall changes in quality of life, and whether patients with a good prognosis would show less negative changes in quality of life than patients with a poor prognosis. The overall observed changes in quality of life were surprisingly small over a 3-4 year period. This study shows that prognoses do predict changes in social functioning and in some physical aspects of quality of life over a 3-4 year period after the assessment of prognoses by physicians. Prognoses did not, however, predict changes in psychological aspects of quality of life.

Assessing the consistency of psychometric properties of the HRQoL scales within the EORTC QLQ-C30 across populations by means of the Mokken Scaling Model

This paper demonstrates how the Mokken Scaling Model and other statistical tools may be useful in assessing the consistency of psychometric properties of health-related quality of life (HRQoL) scales across various populations. The main focus is the psychometric performance of the scales proposed for the EORTC QLQ-C30 in seven patient groups totalling more than 2,000 cancer patients. All scales performed satisfactorily in the total sample with the exception of the role functioning and cognitive functioning scales, which failed in terms of reliability and item discriminant validity. The descriptive statistics for the scales show that several of them, particularly those that build upon only two items, have discrimination problems at the extremes, visible in the high percentages at the maximum or the minimum observed values. The scalability analysis in the subsamples showed that the essential assumption in the Mokken Scaling Model of equal item step order does not hold for the cognitive functioning, emotional functioning and physical functioning scales. We conclude that the Mokken Scaling Model is well suited to the purpose of examining the generalizability of HRQoL scales across subpopulations although a global statistical test of the fit of the measurement model is not available.

Prognostic factors and survival in a heterogeneous sample of cancer patients

This study examines the prognostic value of clinical assessments, including a 3-fold classification of cancer patients by treatment intention. It is based upon a sample of 253 patients with different cancer diagnoses who filled out a 108-item questionnaire. Cox regression analysis (the proportional hazards model) was used to analyse the relationship of the three groups of covariates (clinical, demographic and psychosocial) with survival. The univariate analysis showed that several clinical, demographic and psychosocial covariates are significantly related to survival. The study located two main prognostic factors: the 3-fold classification by treatment intention being the most important one, followed by physical functioning which may be seen as a proxy for performance status. Several additional covariates including psychosocial ones were related to survival when considered separately. However, their effects disappeared when controlling for treatment intention and physical functioning. Thus, the additional psychosocial covariates did no add to the prognostic value of the model.

Quality of life of cancer patients with different prognoses

This study examines differences and similarities in the quality of life of 253 cancer patients with good, medium and poor prognoses. Our main hypothesis was that patients with a good prognosis will experience a higher quality of life than patients with a medium or poor prognosis. A multivariate analysis of covariance of eight quality of life scales was performed with prognosis as a factor and with age, sex, and the duration of the illness as covariates. Significant main effects of prognosis were found for the general QOL-scale and for physical aspects of quality of life. There were, however, only marginal and non-significant effects of prognosis groups on social and psychological functioning. A final multivariate analysis confirms earlier findings that performance status shows a weak but significant relationship with the psychological functioning. Thus, the physical condition of the patient at the time of measurement seems to have some influence on the psychological functioning, whereas the severity of the disease as indicated by the classification into prognosis groups does not. These results question the general attitude that seriously ill cancer patients have reduced social and psychological well-being. An alternative interpretation is that the scales used to measure psychological aspects of quality of life are inadequately sensitive.

Testing the EORTC Quality of Life Questionnaire on cancer patients with heterogeneous diagnoses

This study aimed to contribute to the validation of the 30-item Quality of Life Questionnaire developed by the European Organization for Research and Treatment of Cancer Study Group (EORTC QLQ-C30). The sample consisted of 177 cancer patients with heterogeneous diagnoses. A series of scales representing various dimensions of quality of life were tested, including those proposed by the EORTC Study Group. Mokken's non-parametric latent trait model for unidimensional scaling was used as the basic scaling procedure. This model gives coefficients of scalability in addition to reliability coefficients. In terms of scalability measured by Loevinger's H, all EORTC Study Group scales, except the cognitive functioning scale were found to be quite satisfactory. The cognitive functioning scale and the role functioning scale were below the satisfactory level in terms of reliability (internal consistency). In total, our study strengthens the external validity of the EORTC QLQ-C30 and confirms that it may be used on cancer patients with various diagnoses.