Changes in vital signs as predictors of bacterial infection in home care: a multi-center prospective cohort study

Machine learning-based clinical decision support for infection risk prediction

Background

Healthcare-associated infection (HAI) remains a significant risk for hospitalized patients and a challenging burden for the healthcare system. This study presents a clinical decision support tool that can be used in clinical workflows to proactively engage secondary assessments of pre-symptomatic and at-risk infection patients, thereby enabling earlier diagnosis and treatment.

Methods

This study applies machine learning, specifically ensemble-based boosted decision trees, on large retrospective hospital datasets to develop an infection risk score that predicts infection before obvious symptoms present. We extracted a stratified machine learning dataset of 36,782 healthcare-associated infection patients. The model leveraged vital signs, laboratory measurements and demographics to predict HAI before clinical suspicion, defined as the order of a microbiology test or administration of antibiotics.

Results

Our best performing infection risk model achieves a cross-validated AUC of 0.88 at 1 h before clinical suspicion and maintains an AUC >0.85 for 48 h before suspicion by aggregating information across demographics and a set of 163 vital signs and laboratory measurements. A second model trained on a reduced feature space comprising demographics and the 36 most frequently measured vital signs and laboratory measurements can still achieve an AUC of 0.86 at 1 h before clinical suspicion. These results compare favorably against using temperature alone and clinical rules such as the quick sequential organ failure assessment (qSOFA) score. Along with the performance results, we also provide an analysis of model interpretability via feature importance rankings.

Conclusion

The predictive model aggregates information from multiple physiological parameters such as vital signs and laboratory measurements to provide a continuous risk score of infection that can be deployed in hospitals to provide advance warning of patient deterioration.

Unresolved Palliative Care Needs of Elderly Non-Cancer Patients at Home: A Multicenter Prospective Study

INTRODUCTION/OBJECTIVES

There is growing consensus on the benefits of initiating palliative care early in the disease trajectory; however, palliative care needs for non-cancer patients remain to be elucidated. We investigated the trajectory of unresolved palliative care needs of non-cancer patients at home and explored associated factors.

METHODS

We conducted a multicenter prospective cohort study of elderly non-cancer patients at home in Japan between Jan 2020 and Dec 2020. Physicians assessed their palliative care needs using the Integrated Palliative Care Outcome Scale (IPOS). Unresolved palliative care needs were defined as IPOS symptoms above 2 (moderate).

RESULTS

In total, 785 patients were enrolled. The most frequent unresolved palliative care needs at enrollment were poor mobility (n = 438, 55.8%), followed by weakness/lack of energy (n = 181, 23.1%) and poor appetite (n = 160, 20.4%). Multivariate logistic regression analysis revealed that female and musculoskeletal disease were significantly positively associated with pain at starting home visits (OR = 1.89, P = .015; OR = 2.69, P = .005). In addition, neurological diseases were significantly positively associated with constipation and poor mobility 3 months after starting home visits (OR = 3.75, P = .047; OR = 3.04, P = .009).

CONCLUSIONS

The order of the prevalence of unresolved palliative care needs may remain relatively stable over time, even for those receiving home-based palliative care services. We identified several specific diseases and conditions that were significantly associated with unresolved palliative care needs.

Exploration of the impact of the COVID-19 pandemic on the mental health of home health care workers in Japan: a multicenter cross-sectional web-based survey

Background

The COVID-19 pandemic has caused home health care workers (home-HCWs) to experience anxiety. The mental health of home-HCWs and related factors during the COVID-19 pandemic have not been clarified; therefore, we aimed to investigate the status and associated factors of fear of COVID-19 infection, anxiety, and depression among home-HCWs in Japan.

Methods

We conducted a multicenter cross-sectional web-based anonymous survey of home-HCWs in August 2021, during the fifth wave of the pandemic in Japan. We surveyed members of facilities that provided home visit services during the COVID-19 pandemic. We measured the Japanese version of the Fear of COVID-19 scale (FCV-19S-J) and the Hospital Anxiety and Depression scale (HADS) as objective variables, and the Japanese version of the Assessment of Interprofessional Team Collaboration Scale-II (J-AITCS-II) as an explanatory variable.

Results

A total of 328 members of 37 facilities responded to the survey, and we ultimately analyzed 311 participants. The most frequent occupation was nurse (32.8%), followed by doctor (24.8%) and medical office staff (18.0%). The mean score of the FCV-19S-J was 16.5 ± 5.0 (7.0 – 31.0), and the prevalences of definitive anxiety and depression were 7.4% and 15.7%, respectively. Multivariate regression analysis revealed that the J-AITCS-II teamwork subscale was significantly negatively associated with FCV-19S-J, HADS-anxiety, and HADS-depression (β = -0.171, p = 0.004; β = -0.151, p = 0.012; β = -0.225, p < 0.001, respectively). Medical office staff showed significant positive associations with FCV-19S-J and HADS-depression (β = 0.219, p = 0.005; β = 0.201, p = 0.009, respectively), and medical social workers with HADS-anxiety and HADS-depression (β = -0.166, p = 0.011; β = -0.214, p < 0.001, respectively) compared with doctors. The unmet support need for expert lectures on COVID-19 was significantly positively associated with FCV-19S-J (β = 0.131, p = 0.048), and the unmet support need for support systems for psychological stress and emotional exhaustion was significantly positively associated with HADS-anxiety (β = 0.141, p = 0.022).

Conclusions

Fear of COVID-19 infection and depression of nurses, medical office staff, and other occupations was significantly higher than those of doctors. These findings suggest that non-physicians were more likely to be fearful and depressed during the COVID-19 pandemic; thus, it is necessary to tailor mental health support based on occupation in the home care setting.

Changes in home visit utilization during the COVID-19 pandemic: a multicenter cross-sectional web-based survey

Objective

Home care is one of the essential community health care services; thus, identifying changes of home care utilization before and during the COVID-19 pandemic would be useful for researchers and policymaker to reconsider the home care system, the support needed for home care staff, and the collaborative system with hospitals in the COVID-19 era. We conducted a multicenter cross-sectional web-based anonymous survey of the directors of home visit facilities in Japan in August 2021.

Results

A total of 33 participants from 37 facilities responded to the survey. The number of patients dying at home and newly requested home visits increased during the COVID-19 pandemic (74.2%, 71.0%). One possible reason was the restricted visitation of inpatient facilities (93.5%). The underlying disease that the largest number of participants perceived as having increased compared with before the COVID-19 pandemic was cancer (51.6%). There were no significant differences in being in a rural area or the number of doctors in perceived changes in home visit utilization. Our study indicated that the director of home visit facilities thought the number of patients dying at home and newly requested home visits had increased compared with before the COVID-19 pandemic.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13104-022-06128-7.

Body Temperature Is Associated With Cognitive Performance in Older Adults With and Without Mild Cognitive Impairment: A Cross-sectional Analysis

Wearable devices for remote and continuous health monitoring in older populations frequently include sensors for body temperature measurements (i.e., skin and core body temperatures). Healthy aging is associated with core body temperatures that are in the lower range of age-related normal values (36.3 ± 0.6°C, oral temperature), while patients with Alzheimer's disease (AD) exhibit core body temperatures above normal values (up to 0.2°C). However, the relation of body temperature measures with neurocognitive health in older adults remains unknown. This study aimed to explore the association of body temperature with cognitive performance in older adults with and without mild cognitive impairment (MCI). Eighty community-dwelling older adults (≥65 years) participated, of which 54 participants were cognitively healthy and 26 participants met the criteria for MCI. Skin temperatures at the rib cage and the scapula were measured in the laboratory (single-point measurement) and neuropsychological tests were conducted to assess general cognitive performance, episodic memory, verbal fluency, executive function, and processing speed. In a subgroup (n = 15, nine healthy, six MCI), skin and core body temperatures were measured continuously during 12 h of habitual daily activities (long-term measurement). Spearman's partial correlation analyses, controlled for age, revealed that lower median body temperature and higher peak-to-peak body temperature amplitude was associated with better general cognitive performance and with better performance in specific domains of cognition; [e.g., rib median skin temperature (single-point) vs. processing speed: rs = 0.33, p = 0.002; rib median skin temperature (long-term) vs. executive function: rs = 0.56, p = 0.023; and peak-to-peak core body temperature amplitude (long-term) vs. episodic memory: rs = 0.51, p = 0.032]. Additionally, cognitively healthy older adults showed lower median body temperature and higher peak-to-peak body temperature amplitude compared to older adults with MCI (e.g., rib median skin temperature, single-point: p = 0.035, r = 0.20). We conclude that both skin and core body temperature measures are potential early biomarkers of cognitive decline and preclinical symptoms of MCI/AD. It may therefore be promising to integrate body temperature measures into multi-parameter systems for the remote and continuous monitoring of neurocognitive health in older adults.

A Multicenter Cohort Study to Explore Differentiating Factors between Tumor Fever and Infection among Advanced Cancer Patients

Background: Tumor fever and infection are common febrile etiologies among advanced cancer patients. To date, only few studies have been conducted to differentiate between tumor fever and infections. Objective: This study aimed to identify discriminating factors that provide rapid results and are feasible and minimally invasive for discriminating between tumor fever and infection in advanced cancer patients. Methods: This is a retrospective cohort study. Advanced cancer patients with clinically diagnosed tumor fever or infection, who received medical treatment from palliative care specialists in 10 nationwide Japanese hospitals, were consecutively identified during August 2012 and November 2014. The symptoms, physical findings, blood test results at baseline and during fever, imaging findings, and sociodemographic factors of these patients were retrospectively extracted. Results: Thirty-three patients with tumor fever and 72 patients with infection were identified. Their mean age was 68.8 years, 68 (64.8%) were men, and the median palliative performance status (PPS) was 50. Statistically significant factors predicting tumor fever by logistic regression analysis were as follows: deterioration of PPS (odds ratio, 0.078), shaking chills during fever (0.067), and change from baseline data of neutrophil/lymphocyte ratio of ≥5 (0.14). Conclusions: Shaking chills during fever, and changes from baseline of performance status and white blood cell differentiation can be useful to differentiate between tumor fever and infection among advanced cancer patients. Further confirmatory studies are needed.