Perceived risk of institutionalization, hospitalization, and death in oldest old primary care patients

The role of geriatric syndromes in predicting unplanned hospitalizations: a population-based study using Minimum Data Set for Home Care

BACKGROUND

The predictive accuracies of screening instruments for identifying home-dwelling old people at risk of hospitalization have ranged from poor to moderate, particularly among the oldest persons. This study aimed to identify variables that could improve the accuracy of a Minimum Data Set for Home Care (MDS-HC) based algorithm, the Detection of Indicators and Vulnerabilities for Emergency Room Trips (DIVERT) Scale, in classifying home care clients' risk for unplanned hospitalization.

METHODS

In this register-based retrospective study, factors associated with hospitalization among home care clients aged ≥ 80 years in the City of Tampere, Finland, were analyzed by linking MDS-HC assessments with hospital discharge records. MDS-HC determinants associated with hospitalization within 180 days after the assessment were analyzed for clients at low (DIVERT 1), moderate (DIVERT 2-3) and high (DIVERT 4-6) risk of hospitalization. Then, two new variables were selected to supplement the DIVERT algorithm. Finally, area under curve (AUC) values of the original and modified DIVERT scales were determined using the data of MDS-HC assessments of all home care clients in the City of Tampere to examine if addition of the variables related to the oldest age groups improved the accuracy of DIVERT.

RESULTS

Of home care clients aged ≥ 80 years, 1,291 (65.4%) were hospitalized at least once during the two-year study period. Unplanned hospitalization occurred following 15.9%, 22.8%, and 33.9% MDS-HC assessments with DIVERT group 1, 2-3 and 4-6, respectively. Infectious diseases were the most common diagnosis within each DIVERT groups. Many MDS-HC variables not included in the DIVERT algorithm were associated with hospitalization, including e.g. poor self-rated health and old fracture (other than hip fracture) (p 0.001) in DIVERT 1; impaired cognition and decision-making, urinary incontinence, unstable walking and fear of falling (p < 0.001) in DIVERT 2-3; and urinary incontinence, poor self-rated health (p < 0.001), and decreased social interaction (p 0.001) in DIVERT 4-6. Adding impaired cognition and urinary incontinence to the DIVERT algorithm improved sensitivity but not accuracy (AUC 0.64 (95% CI 0.62-0.65) vs. 0.62 (0.60-0.64) of the original DIVERT). More admissions occurred among the clients with higher scores in the modified than in the original DIVERT scale.

CONCLUSIONS

Certain geriatric syndromes and diagnosis groups were associated with unplanned hospitalization among home care clients at low or moderate risk level of hospitalization. However, the predictive accuracy of the DIVERT could not be improved. In a complex clinical context of home care clients, more important than existence of a set of risk factors related to an algorithm may be the various individual combinations of risk factors.

Classifying home care clients' risk of unplanned hospitalization with the resident assessment instrument

Aim

The aim was to identify tools for classifying the risk of unplanned hospitalization among home care clients utilizing the Resident Assessment Instrument-Home Care (RAI-HC).

Findings

The Detection of Indicators and Vulnerabilities for Emergency Room Trips (DIVERT) Scale predicts unplanned hospitalizations in home care clients. In the oldest age groups, however, it works poorly.

Message

DIVERT Scale can be used for identifying high-risk home care clients needing urgent care planning to prevent unplanned hospital admissions and their potential adverse consequences. Clients scoring high in the scale and experiencing the outcome earlier than others, should be the primary group for more detailed assessment.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41999-022-00665-x.

Purpose

To identify predictive case finding tools for classifying the risk of unplanned hospitalization among home care clients utilizing the Resident Assessment Instrument-Home Care (RAI-HC), with special interest in the Detection of Indicators and Vulnerabilities for Emergency Room Trips (DIVERT) Scale.

Methods

A register-based, retrospective study based on the RAI-HC assessments of 3,091 home care clients (mean age 80.9 years) in the City of Tampere, Finland, linked with hospital discharge records. The outcome was an unplanned hospitalization within 180 days after RAI-HC assessment. The Area Under the Curve (AUC) and the sensitivity and specificity were determined for the RAI-HC scales: DIVERT, Activities of Daily Living Hierarchy (ADLh), Cognitive Performance Scale (CPS), Changes in Health, End-Stage Diseases, Signs, and Symptoms Scale (CHESS), and Method for Assigning Priority Levels (MAPLe).

Results

Altogether 3091 home care clients had a total of 7744 RAI-HC assessments, of which 1658 (21.4%) were followed by an unplanned hospitalization. The DIVERT Scale had an AUC of 0.62 (95% confidence interval 0.61–0.64) when all assessments were taken into account, but its value was poorer in the older age groups (< 70 years: 0.71 (0.65–0.77), 70–79 years: 0.66 (0.62–0.69), 80–89 years: 0.60 (0.58–0.62), ≥ 90 years: 0.59 (0.56–0.63)). AUCs for the other scales were poorer than those of DIVERT, with CHESS nearest to DIVERT. Time to hospitalization after assessment was shorter in higher DIVERT classes.

Conclusion

The DIVERT Scale offers an approach to predicting unplanned hospitalization, especially among younger home care clients. Clients scoring high in the DIVERT algorithm were at the greatest risk of unplanned hospitalization and more likely to experience the outcome earlier than others.

Supplementary Information

The online version contains supplementary material available at 10.1007/s41999-022-00665-x.

Validation of the Portuguese Version of the Risk Instrument for Screening in the Community (RISC) Among Older Patients in Primary Care in Northern Portugal

Background: Aging is associated with an increase in adverse health outcomes for older people. Short screening instruments that easily and quickly identify those at highest risk can enable decision-makers to anticipate future needs, allocate scarce resources and act to minimize risk. The Risk Instrument for Screening in the Community (RISC) is a brief (2-5 min) Likert scale that scores one-year risk of institutionalization, hospitalization and death from low (1/5) to severe (5/5). Objectives: To externally validate the RISC, scored by general practitioners (GP's), in primary care in Northern Portugal. Methods: The RISC was translated and culturally adapted to Portuguese. A cohort of 457 older adults (aged ≥65) under active follow-up with their GP's were screened. Outcomes at one-year were recorded. Accuracy was determined from the area under the curve (AUC) of receiver operating curve analysis. Results: The mean age of participants was 75.2 years; 57% were female. The proportion identified as being at maximum risk (RISC scores of 3-5/5) of institutionalization, hospitalization and death, were 14.9, 52.4 and 38.4%, respectively. At follow-up 2% (10/431) were institutionalized, 18.6% (84/451) were hospitalized and 3% (14/456) died. Those who were institutionalized (p = 0.021), hospitalized (p = 0.012) or dead (p < 0.001) at one-year were significantly older. Those living alone were more likely to be institutionalized (p = 0.007). The RISC showed fair accuracy in predicting hospitalization (AUC of 0.62 [95% CI: 0.55-0.69]) and good accuracy for Institutionalization (AUC of 0.79 [95% CI: 0.62-0.96]) and death (AUC of 0.77 [95% CI: 0.65-0.88]). Conclusions: The Portuguese version of the RISC accurately predicted institutionalization and death at one-year but like most short screens was less able to predict hospitalization. Given its brevity, the RISC is useful for quickly identifying and stratifying those at increased risk in primary care.

Transforming a Patient Registry Into a Customized Data Set for the Advanced Statistical Analysis of Health Risk Factors and for Medication-Related Hospitalization Research: Retrospective Hospital Patient Registry Study

BACKGROUND

Hospital patient registries provide substantial longitudinal data sets describing the clinical and medical health statuses of inpatients and their pharmacological prescriptions. Despite the multiple advantages of routinely collecting multidimensional longitudinal data, those data sets are rarely suitable for advanced statistical analysis and they require customization and synthesis.

OBJECTIVE

The aim of this study was to describe the methods used to transform and synthesize a raw, multidimensional, hospital patient registry data set into an exploitable database for the further investigation of risk profiles and predictive and survival health outcomes among polymorbid, polymedicated, older inpatients in relation to their medicine prescriptions at hospital discharge.

METHODS

A raw, multidimensional data set from a public hospital was extracted from the hospital registry in a CSV (.csv) file and imported into the R statistical package for cleaning, customization, and synthesis. Patients fulfilling the criteria for inclusion were home-dwelling, polymedicated, older adults with multiple chronic conditions aged ≥65 who became hospitalized. The patient data set covered 140 variables from 20,422 hospitalizations of polymedicated, home-dwelling older adults from 2015 to 2018. Each variable, according to type, was explored and computed to describe distributions, missing values, and associations. Different clustering methods, expert opinion, recoding, and missing-value techniques were used to customize and synthesize these multidimensional data sets.

RESULTS

Sociodemographic data showed no missing values. Average age, hospital length of stay, and frequency of hospitalization were computed. Discharge details were recoded and summarized. Clinical data were cleaned up and best practices for managing missing values were applied. Seven clusters of medical diagnoses, surgical interventions, somatic, cognitive, and medicines data were extracted using empirical and statistical best practices, with each presenting the health status of the patients included in it as accurately as possible. Medical, comorbidity, and drug data were recoded and summarized.

CONCLUSIONS

A cleaner, better-structured data set was obtained, combining empirical and best-practice statistical approaches. The overall strategy delivered an exploitable, population-based database suitable for an advanced analysis of the descriptive, predictive, and survival statistics relating to polymedicated, home-dwelling older adults admitted as inpatients. More research is needed to develop best practices for customizing and synthesizing large, multidimensional, population-based registries.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

RR2-10.1136/bmjopen-2019-030030.