Extended hospital stays with increasing age: the impact of an acute geriatric unit

Predicting Postoperative Hospital Stays Using Nursing Narratives and the Reverse Time Attention (RETAIN) Model: Retrospective Cohort Study

Background

Nursing narratives are an intriguing feature in the prediction of short-term clinical outcomes. However, it is unclear which nursing narratives significantly impact the prediction of postoperative length of stay (LOS) in deep learning models.

Objective

Therefore, we applied the Reverse Time Attention (RETAIN) model to predict LOS, entering nursing narratives as the main input.

Methods

A total of 354 patients who underwent ovarian cancer surgery at the Seoul National University Bundang Hospital from 2014 to 2020 were retrospectively enrolled. Nursing narratives collected within 3 postoperative days were used to predict prolonged LOS (≥10 days). The physician's assessment was conducted based on a retrospective review of the physician's note within the same period of the data model used.

Results

The model performed better than the physician's assessment (area under the receiver operating curve of 0.81 vs 0.58; P=.02). Nursing narratives entered on the first day were the most influential predictors in prolonged LOS. The likelihood of prolonged LOS increased if the physician had to check the patient often and if the patient received intravenous fluids or intravenous patient-controlled analgesia late.

Conclusions

The use of the RETAIN model on nursing narratives predicted postoperative LOS effectively for patients who underwent ovarian cancer surgery. These findings suggest that accurate and interpretable deep learning information obtained shortly after surgery may accurately predict prolonged LOS.

Survival and neurological outcome in the elderly after in-hospital cardiac arrest

BACKGROUND

There have been few studies of the outcome in elderly patients who have suffered in-hospital cardiac arrest (IHCA) and the association between cardiac arrest characteristics and survival.

AIM

The aim of this large observational study was to investigate the survival and neurological outcome in the elderly after IHCA, and to identify which factors were associated with survival.

METHODS

We investigated elderly IHCA patients (≥70years of age) who were registered in the Swedish Cardiopulmonary Resuscitation Registry 2007-2015. For descriptive purposes, the patients were grouped according to age (70-79, 80-89, and ≥90years). Predictors of 30-day survival were identified using multivariable analysis.

RESULTS

Altogether, 11,396 patients were included in the study. Thirty-day survival was 28% for patients aged 70-79 years, 20% for patients aged 80-89 years, and 14% for patients aged ≥90years. Factors associated with higher survival were: patients with an initially shockable rhythm, IHCA at an ECG-monitored location, IHCA was witnessed, IHCA during daytime (8 a.m.-8 p.m.), and an etiology of arrhythmia. A lower survival was associated with a history of heart failure, respiratory insufficiency, renal dysfunction and with an etiology of acute pulmonary oedema. Patients over 90 years of age with VF/VT as initial rhythm had a 41% survival rate. We found a trend indicating a less aggressive care with increasing age during cardiac arrest (fewer intubations, and less use of adrenalin and anti-arrhythmic drugs) but there was no association between age and delay in starting cardiopulmonary resuscitation (CPR). In survivors, there was no significant association between age and a favourable neurological outcome (CPC score: 1-2) (92%, 93%, and 88% in the three age groups, respectively).

CONCLUSIONS

Increasing age among the elderly is associated with a lower 30-day survival after IHCA. Less aggressive treatment and a worse risk profile might contribute to these findings. Relatively high survival rates among certain subgroups suggest that discussions about advanced directives should be individualized. Most survivors have good neurological outcome, even patients over 90 years of age.

Should There Be an Age Split for Stroke DRGs? Analysing a Large Clinical Data Set of a Principal Teaching Hospital over a Five-Year Period

The aim of this study was to analyse the inpatient statistics collection relating to stroke patients admitted to a major teaching hospital, with particular reference to length of stay, and to assess the adequacy of the diagnosis related group (DRG) as a predictor of length of stay. The study subjects were selected by DRG to identify all stroke inpatients admitted and discharged between 1 July 1995 and 30 June 2000. There were 1365 stroke discharges (half of whom were over 75 years of age at discharge) over the period of the study. The median length of stay was 8 days, and 67% of the subjects experienced complications and/or comorbidities. Age was significantly associated with increased length of stay of stroke patients, independent of complications or comorbidities.

These findings raise the question of whether casemix-based funding should be based solely on DRGs for complicated conditions such as stroke, or whether additional measures such as age should be used for funding allocation. This study provides a model that health information managers and other researchers could use to analyse inpatient statistics collections at state, territory or national levels.

Adverse outcomes following hospitalization in acutely ill older patients

Background

The longitudinal outcomes of patients admitted to acute care for elders units (ACE) are mixed. We studied the associations between socio-demographic and functional measures with hospital length of stay (LOS), and which variables predicted adverse events (non-independent living, readmission, death) 3 and 6 months later.

Methods

Prospective cohort study of community-living, medical patients age 75 or over admitted to ACE at a teaching hospital.

Results

The population included 147 subjects, median LOS of 9 days (interquartile range 5–15 days). All returned home/community after hospitalization. Just prior to discharge, baseline timed up and go test (TUG, P < 0.001), bipedal stance balance (P = 0.001), and clinical frailty scale scores (P = 0.02) predicted LOS, with TUG as the only independent predictor (P < 0.001) in multiple regression analysis. By 3 months, 59.9% of subjects remained free of an adverse event, and by 6 months, 49.0% were event free. The 3 and 6-month mortality was 10.2% and 12.9% respectively. Almost one-third of subjects had developed an adverse event by 6 months, with the highest risk within the first 3 months post discharge. An abnormal TUG score was associated with increased adjusted hazard ratio [HR] 1.28, 95% confidence interval [CI] 1.03 to 1.59, P = 0.03. A higher FMMSE score (adjusted HR 0.89, 95% CI 0.82 to 0.96, P = 0.003) and independent living before hospitalization (adjusted HR 0.42, 95% CI 0.21 to 0.84, P = 0.01) were associated with reduced risk of adverse outcome.

Conclusion

Some ACE patients demonstrate further functional decline following hospitalization, resulting in loss of independence, repeat hospitalization, or death. Abnormal TUG is associated with prolonged LOS and future adverse outcomes.

Geriatric syndromes as predictors of adverse outcomes of hospitalization

AIMS

The aim of this study was to determine the association between the common geriatric syndromes and predefined adverse outcomes of hospitalization and to identify the most important independent predictors of adverse outcomes using information gained within 24 h of admission in older general medical patients.

METHODS

A prospective longitudinal cohort study of patients aged > or =75 years admitted to the rapid assessment medical unit in a teaching hospital was carried out. The role of geriatric syndromes in predicting outcomes was examined in univariate and multivariate models. The outcome measures were (i) length of hospital stay (LOS) of 28 days or more, (ii) institutionalization or change in residential care status to a more dependent category at discharge or during 3 months post-discharge, (iii) unplanned readmissions during 3 months and (iv) mortality in hospital or 3 months post-discharge.

RESULTS

The presence of geriatric syndromes was significantly associated with increased LOS and institutionalization or change in residential care status to a more dependent category. The factors most predictive of these outcomes were impaired pre-admission functional status in activities of daily living, recurrent falls, urinary incontinence and supported living arrangements. The geriatric syndromes appeared less important in predicting unplanned readmission and death.

CONCLUSION

The presence of geriatric syndromes in older general medical patients is an important determinant of adverse outcomes of hospitalization, particularly of LOS and admission to residential care. The predictors most useful for screening patients for these outcomes, within 24 h of admission, appear to be the presence of certain pre-existing geriatric syndromes before admission.

Reduced utilisation of specialist care among elderly cancer patients: a randomised study of a primary healthcare intervention

The aim of this study was to evaluate the effect of an individual support (IS) intervention including intensified primary healthcare on the utilisation of specialist care among cancer patients, and to investigate if such an effect was modified by the patient's age (less than 70 years or 70 years and more). Newly diagnosed cancer patients (n=416) were randomised between the intervention and a control condition, and data were collected on the utilisation of specialist care within 3 months from inclusion. Intensified primary healthcare comprised extended information from the specialist clinics, and education and supervision in cancer care for general practitioners (GPs) and home-care nurses. The support given also included interventions designed to diminish problems of weight loss and psychological distress. The intervention reduced the number of admissions (NoA) and the days of hospitalisation (DoH) after adjustment for weight loss and psychological distress, but only for older patients. Older patients randomised to the intervention (n=82) experienced 393 fewer DoH than the older control patients (n=79). In addition, the proportion of older patients in the IS group who utilised acute specialist care was smaller compared with older control patients group. The conclusion is that older cancer patients' utilisation of specialist care may be reduced by intensified primary healthcare services.

Are elderly patients with obstructive airway disease being prematurely discharged?

Despite the temporal trend toward decreasing length of hospital stay for all medical conditions in North America, the effect of different lengths of hospitalization on short-term outcomes such as readmission or mortality has not been well studied. However, there is growing concern that very short stays in hospital may result in premature discharges, which may lead to worse outcomes for patients. We conducted a population-based study of elderly patients with obstructive airway disease in Ontario, Canada to test the hypothesis that very short initial hospital stays increase the short-term risk for readmission and mortality. Using a cohort of 32,384 elderly patients 65 yr of age or older, we compared 15-d rates of readmission and mortality among patients with different lengths of stay. Although patients with hospital stays of less than 4 d were younger and had fewer comorbidities, they were 39% (95% confidence interval [CI], 20% to 61%) more likely to be readmitted and 45% (95% CI, 9% to 92%) more likely to die within 15 d postdischarge compared with those who stayed 4 to 6 d. The risk was highest among patients whose stay was less than or equal to 1 hospital day; they had a 69% (95% CI, 32% to 117%) excess risk of readmission and a 2.08 (95% CI, 1.23 to 3.45) -fold increase in mortality compared with those who stayed in hospital for 2 d. This suggests that some elderly patients with obstructive airway disease may be being prematurely discharged.

Fitting the distributions of length of stay by parametric models

OBJECTIVES

The purpose of this study was to assess the adequacy of three widely used models--Lognormal, Weibull, and Gamma--for describing the distribution of length of stay. This is a fundamental step in the development of outliers resistant (robust) methods for the statistical analysis of this kind of data, where the main objective is to determine measures of average and total resource consumption of groups of patients. Current practice uses several types of trimming rules, many of which are based on the Lognormal model, although theoretical and experimental bases are still insufficient.

METHODS

The three models were adjusted using robust procedures based on M-estimators to approximately 5 million stays grouped by Diagnosis-Related Groups (DRGs). The resulting 3,279 samples were collected in five European countries during 3 years.

RESULTS

Most of the distributions observed could be fitted with one of these models. The descriptions provided by the Gamma and the Weibull models were similar, and the Gamma model could be omitted. The casemix description provided by the Log-normal-Weibull family was, for certain countries, significantly better than the one provided by the single Lognormal model. Often, for a given DRG and a given country, length of stay distributions could be described with the same model during several years. A given DRG, however, usually had to be described by means of different models for different countries.

CONCLUSIONS

Practical and conceptual consequences of the results are discussed. They can be extended to the analyses of other consumption variables used in health services. Statistical procedures for casemix description, including current rules of trimming, should be improved by means of more flexible families of models.