Patient reported symptoms, body temperature and hospital mortality: an observational study in a low resource healthcare environment.

BACKGROUND: Fever is a common presenting complaint of patients, especially in sub-Saharan Africa. Although most medical authorities consider fever to be synonymous with an elevated body temperature the relationship of the complaint of fever made by patients to temperature has not been well defined. AIM: This study examined the relationship of the complaint of fever to temperature on and after admission and in-hospital mortality. METHOD: Observational study in a low-resource Ugandan mission hospital. RESULTS: Out of 2122 alert patients admitted between 9 August 2016 and 5 January 2018, 349 (16.4%) complained of fever: these patients were no more likely to have an abnormal temperature or die in-hospital than those not complaining of fever. Of the 707 alert patients admitted after 1 July 2017, 422 were interviewed in detail about their symptoms: only rigors, feeling intermittently hot and cold, and anorexia were statistically related to the complaint of fever, and only rigors to an admission temperature >38°C. No symptom or sign was associated with a temperature ≤36°C: cold and clammy skin was the only finding associated with in-hospital death. On logistic regression the only independent predictors of mortality were: the National Early Warning Score, impaired mobility on presentation and cold and clammy skin. CONCLUSION: In this study, the term fever used by patients and raised body temperature were not synonymous. Although fever and related symptoms reported by patients are common presenting complaints only the finding of cold and clammy skin was associated with in-hospital mortality.

Should impaired mobility on presentation be a vital sign?

INTRODUCTION: Vital signs are routinely used to assess acutely ill patients, but they do not detect all patients at risk of death. This retrospective multicenter cohort study compares the prediction of death by impaired mobility with age, co-morbidities, and vital sign changes. METHODS: On first assessment, patients from a combined cohort of 9684 Danish and Irish patients and a separate cohort of 1010 Ugandan patients were stratified by impaired mobility on presentation (IMOP), vital sign changes assessed by the National Early Warning Score (NEWS), the Charlson Co-morbidity Index, and age. RESULTS: Fourteen percent of Danish and Irish patients had IMOP compared with 42% of Ugandan patients. The odds ratios of IMOP for 7-day mortality were similar for both cohorts (i.e. 11.8, 95% CI 5.8-24.0 for Ugandan patients versus 6.7, 95% CI 5.0-9.0 for Danish and Irish patients). Univariate analysis of Ugandan patients showed that none of the parameters tested (i.e. low blood pressure, pulse, elevated respiratory rate, hypothermia, low oxygen saturation, old age, and coma) had a statistically higher odds ratio for either 7-day mortality than IMOP. Multivariate logistic regression analysis of Danish and Irish patients also showed that none of these parameters or the Charlson Co-morbidity Index had a statistically higher odds ratio than IMOP for either 7-day or 30-day mortality. CONCLUSION: Immobility on presentation is a vital sign and predicts mortality for acutely ill patients independently of the traditional vital signs, age, and co-morbidities.

Does how the patient feels matter? A prospective observational study of the outcome of acutely ill medical patients who feel their condition has improved on their first re-assessment after admission to hospital.

BACKGROUND: Although asking how a patient feels is the first enquiry most clinicians make the value of the answer has never been examined in acutely ill patients. METHODS: Prospective observational study that compared the predictive value of how well acutely ill medical patients felt after admission to a resource poor sub-Saharan hospital with their mental alertness, mobility and vital signs. RESULTS: In total, 403 patients were studied. Patients who felt better when re-assessed 18.0 SD 9.1 h after admission to hospital were less likely to die in hospital (OR 0.18 95% CI 0.08-0.43, P = 0.00001) and more likely to be independent of others at discharge (OR 5.64 95% CI 3.04-10.47, P = 0.00001). Feeling better was an independent predictor of in-hospital death along with vital sign changes and gait stability, and an independent predictor of independence at discharge along with vital sign changes, gait stability and female gender. CONCLUSION: In this patient cohort a subjective feeling of improvement at the first re-assessment after admission to hospital is a powerful independent predictor of reduced in-hospital mortality.

Performance of TOTAL, in medical patients attending a resource-poor hospital in sub-Saharan Africa and a small Irish rural hospital.

BACKGROUND: Recently a very simple, easy to remember early warning score (EWS) dubbed TOTAL has been reported. The score was derived from 309 acutely ill medical patients admitted to a Malawian hospital and awards one point for Tachypnea >30 breaths per minute, one point for Oxygen saturation <90%, two points for a Temperature <35°C, one point for Altered mental status, and one point for Loss of independence as indicated by the inability to stand or walk without help. TOTAL has an area under the receiver operator characteristic curve (AUROC) for death within 72 hours of 78%. METHODS: We compared the performance of the TOTAL score in 849 medical patients attending a resource poor hospital in Uganda and 2935 patients admitted to a small rural hospital in Ireland. RESULTS: TOTAL's AUROC for death within 24 hours was the same in both hospital populations: 85.1% (95% CI 78.6 - 91.6%) for Kitovu Hospital patients and 84.7% (95% CI 77.1 - 92.2%) for Nenagh Hospital patients. CONCLUSION: The discrimination of TOTAL is exactly the same in elderly Irish patients as it is in young African patients. The score is easy to remember, easy to calculate, and works over a broad range of patients.