Optimising the ISAR-HP to screen efficiently for functional decline in older patients.

INTRODUCTION: The Identification of Seniors At Risk-Hospitalised Patients (ISAR-HP) has recently been included in guidelines as a frailty indicator to identify patients for comprehensive geriatric assessment. Previous studies showed that the conventional cut-off score incorrectly classifies a high percentage of patients as high risk. We aimed to optimise the predictive value of ISAR-HP by using different cut-offs in older acutely hospitalised patients. METHODS: A prospective follow-up study was performed in two Dutch hospitals. Acutely hospitalised patients aged ≥ 70 years were included. Demographics, illness severity parameters, geriatric measurements and the ISAR-HP scores were obtained at baseline. The primary outcome was a combined end point of functional decline or mortality during 90-day follow-up. RESULTS: In total 765 acutely hospitalised older patients were included, with a median age of 79 years, of whom 276 (36.1%) experienced functional decline or mortality. The conventional ISAR-HP cut-off of ≥ 2 assigned 432/765 patients (56.5%) as high risk, with a positive predictive value (PPV) of 0.49 (95%CI 0.45-0.54) and a negative predictive value of 0.81 (95%CI 0.76-0.85). Thus, 51% of those whom the ISAR-HP denoted as high risk did not experience the outcome of interest. Raising the cut-off to ≥ 4 assigned 205/765 patients (26.8%) as high risk, with a marginally increased PPV to 0.55 (95%CI 0.48-0.62). CONCLUSION: The ISAR-HP with the conventional cut-off of ≥ 2 incorrectly identifies a large group of patients at high risk for functional decline or mortality and raising the cut-off to 4 only marginally improved performance. Caution is warranted to ensure efficient screening and follow-up interventions.

Predicting adverse health outcomes in older emergency department patients: the APOP study.

BACKGROUND: Older patients experience high rates of adverse outcomes after an emergency department (ED) visit. Early identification of those at high risk could guide preventive interventions and tailored treatment decisions, but available models perform poorly in discriminating those at highest risk. The present study aims to develop and validate a prediction model for functional decline and mortality in older patients presenting to the ED. METHODS: A prospective follow-up study in patients aged ≥ 70, attending the EDs of the LUMC, the Netherlands (derivation) and Alrijne Hospital, the Netherlands (validation) was conducted. A baseline assessment was performed and the main outcome, a composite of functional decline and mortality, was obtained after 90 days of follow-up. RESULTS: In total 751 patients were enrolled in the Leiden University Medical Center of whom 230 patients (30.6%) experienced the composite outcome and 71 patients (9.5%) died. The final model for the composite outcome resulted in an area under the curve (AUC) of 0.73 (95% CI 0.67-0.77) and was experienced in 69% of the patients at highest risk. For mortality the AUC was 0.79 (95% CI 0.73-0.85) and 36% of the patients at highest risk died. External validation in 881 patients of Alrijne Hospital showed an AUC of 0.71 (95% CI 0.67-0.75) for the composite outcome and 0.67 (95% CI 0.60-0.73) for mortality. CONCLUSION: We successfully developed and validated prediction models for 90-day composite outcome and 90-day mortality in older emergency patients. The benefits for patient management by implementing these models with preventive interventions have to be investigated.

PP167. A process evaluation of an innovative implementation strategy of the Dutch guidelines on hypertensive disorders in pregnancy using a computerized decision support system.

INTRODUCTION: Hypertensive disorders in pregnancy remain the leading cause of maternal mortality in the Netherlands. The Dutch Society of Obstetrics and Gynecology (NVOG) has developed evidence-based guidelines on the management of hypertension in pregnancy. Previous studies showed a low adherence rate to other NVOG guidelines and a large variation in usual care in different hospitals. In the BIG CHANGE trial an innovative implementation strategy of the NVOG guidelines on hypertension using a web-based application (BOS, by Giant Soft, Leeuwarden, The Netherlands) was compared to a common strategy of professional audit and feedback. In this study a process evaluation of BOS has been done, analyzing its efficiency, barriers and formulate improvement points. OBJECTIVES: Gynecologists, residents and clinical midwives from seven hospitals using BOS were asked to fill in the questionnaire. METHODS: A questionnaire was developed on the following items: efficiency, barriers and improvement. RESULTS: Thirty four completed questionnaires useful for analysis. 63.6% of the respondent also consulted the NVOG guideline or local protocol, mainly for confirmation of information, background information, medication. Technical problems were found in 44.1%. Positive opinions on user friendliness varied from 73.5% to 100%. No significant difference was found between the user frequency of BOS compared to the NVOG guidelines or local protocol, or between the time needed to consult them. Improvements mentioned by the respondents were mainly regarding the lay-out. Most respondents (85.3%) found it useful to make a computer based support system for other guidelines and 79.4% would also use this. CONCLUSION: BOS is regarded suitable as an instrument for implementing guidelines and respondents find it useful to develop it for other guidelines as well. Technical problems and poor implementation are important areas of improvement.