Exploring Trajectories of Health Care Utilization Before and After Surgery

Automated-detection of risky alcohol use prior to surgery using natural language processing

BACKGROUND

Preoperative risky alcohol use is one of the most common surgical risk factors. Accurate and early identification of risky alcohol use could enhance surgical safety. Artificial Intelligence-based approaches, such as natural language processing (NLP), provide an innovative method to identify alcohol-related risks from patients' electronic health records (EHR) before surgery.

METHODS

Clinical notes (n = 53,629) from pre-operative patients in a tertiary care facility were analyzed for evidence of risky alcohol use and alcohol use disorder. One hundred of these records were reviewed by experts and labeled for comparison. A rule-based NLP model was built, and we assessed the clinical notes for the entire population. Additionally, we assessed each record for the presence or absence of alcohol-related International Classification of Diseases (ICD) diagnosis codes as an additional comparator.

RESULTS

NLP correctly identified 87% of the human-labeled patients classified with risky alcohol use. In contrast, diagnosis codes alone correctly identified only 29% of these patients. In terms of specificity, NLP correctly identified 84% of the non-risky cohort, while diagnosis codes correctly identified 90% of this cohort. In the analysis of the full dataset, the NLP-based approach identified three times more patients with risky alcohol use than ICD codes.

CONCLUSIONS

NLP, an artificial intelligence-based approach, efficiently and accurately identifies alcohol-related risk in patients' EHRs. This approach could supplement other alcohol screening tools to identify patients in need of intervention, treatment, and/or postoperative withdrawal prophylaxis. Alcohol-related ICD diagnosis had limited utility relative to NLP, which extracts richer information within clinical notes to classify patients.

Social inequalities in trajectories of contacts with the healthcare system in adolescence and young adulthood

BACKGROUND

Understanding of healthcare utilization of different populations is useful for prevention and prioritization of healthcare resources. This study aims to identify populations following different trajectories of contacts with the healthcare system and to describe social inequalities between the groups.

METHODS

Individuals born 1980-2000 in Denmark were linked to national registers. Contacts with somatic hospitals, psychiatric hospitals, general practitioners, and redeemed prescriptions were counted for each year between 16 and 37 years of age. Trajectories of contacts with the four dimensions of healthcare use were identified using group-based multi-trajectory modeling.

RESULTS

Five trajectory groups were identified. One group had low healthcare utilization over time (12% in women; 27% in men). The largest group had low healthcare utilization but more contacts with especially GP (39% in women; 43% in men). A third group had more contacts with most dimensions of the healthcare system (33% in women; 21% in men). The fourth group had many contacts with especially somatic hospitals and GP (7% in women; 4% in men). The fifth group had many contacts especially to psychiatric hospitals (8% in women; 5% in men). Shorter parental education, parental unemployment, family income below the poverty line, and cohabitation with one or no parent was more frequent in the two high utilization groups compared to the lower utilization groups.

CONCLUSION

The observed trajectories of health service use and the social inequalities between trajectory groups highlight that prevention and treatment targeting the entire population will benefit from a complementary focus on social inequalities in health.

Characterizing hospitalization trajectories in the high-need, high-cost population using electronic health record data

High utilization by a minority of patients accounts for a large share of health care costs, but the dynamics of this utilization remain poorly understood. We sought to characterize longitudinal trajectories of hospitalization among adult patients at an academic medical center from 2017 to 2023. Among 3404 patients meeting eligibility criteria, following an initial "rising-risk" period of 3 hospitalizations in 6 months, growth mixture modeling discerned 4 clusters of subsequent hospitalization trajectories: no further utilization, low chronic utilization, persistently high utilization with a slow rate of increase, and persistently high utilization with a fast rate of increase. Baseline factors associated with higher-order hospitalization trajectories included admission to a nonsurgical service, full code status, intensive care unit-level care, opioid administration, discharge home, and comorbid cardiovascular disease, end-stage kidney or liver disease, or cancer. Characterizing hospitalization trajectories and their correlates in this manner lays groundwork for early identification of those most likely to become high-need, high-cost patients.

Yarning about pain: Evaluating communication training for health professionals at persistent pain services in Queensland, Australia

Background

Providing cultural education to health professionals is essential in improving the quality of care and outcomes for Aboriginal and Torres Strait Islander patients. This study reports the evaluation of a novel training workshop used as an intervention to improve communication with Aboriginal and Torres Strait Islander patients of persistent pain services.

Methods

In this single-arm intervention study, health professionals undertook a one-day workshop, which included cultural capability and communication skills training based on a clinical yarning framework. The workshop was delivered across three adult persistent pain clinics in Queensland. At the end of the training, participants completed a retrospective pre/post evaluation questionnaire (5 points Likert scale, 1 = very low to 5 = very high), to rate their perceived importance of communication training, their knowledge, ability and confidence to communicate effectively. Participants also rated their satisfaction with the training and suggested improvements for future trainings.

Results

Fifty-seven health professionals were trained (N = 57/111; 51% participation rate), 51 completed an evaluation questionnaire (n = 51/57; 90% response rate). Significant improvements in the perceived importance of communication training, knowledge, ability and confidence to effectively communicate with Aboriginal and Torres Strait Islander patients were identified (p < 0.001). The greatest increase was in the perceived confidence pre-training mean of 2.96 (SE = 0.11) to the post-training mean of 4.02 (SE = 0.09).

Conclusion

This patient-centred communication training, delivered through a novel model that combines cultural capability and the clinical yarning framework applied to the pain management setting, was highly acceptable and significantly improved participants' perceived competence. This method is transferrable to other health system sectors seeking to train their clinical workforce with culturally sensitive communication skills.

Identifying common patterns of health services use: a longitudinal study of older Swiss adults' care trajectories

BACKGROUND

Population ageing puts pressure on health systems initially designed to handle acute and episodic illnesses. Segmenting an ageing population based on its healthcare utilization may enable policymakers to undertake evidence-based resource planning. We aimed to derive a typology of healthcare utilization trajectories in Swiss older adults.

METHODS

Our work used data from the Lc65 + study, a population-based cohort of individuals aged 65 to 70 years at enrolment. The dimensions of healthcare utilization considered were ambulatory care, emergency care, hospitalizations, professional home care and nursing home stay. We applied the Sequence Analysis framework, within which we quantified the variation between each multidimensional pair of sequences, implemented a clustering procedure that grouped together older persons with similar profiles of health services use, and characterized clusters of individuals using selected baseline covariates.

RESULTS

Healthcare utilization trajectories were analysed for 2271 community-dwelling older adults over a period of 11 years. Six homogeneous subgroups were identified: constant low utilization (83.3% of participants), increased utilization (4.9%), late health deterioration (4.4%), ambulatory care to nursing home (1.5%), early fatal event (3.8%) and high ambulatory care (2.1%). Associations were found between cluster membership and age, sex, household composition, self-perceived health, grip strength measurement, comorbidities, and functional dependency.

CONCLUSIONS

The heterogeneous healthcare utilization profiles can be clustered into six common patterns. Different manifestations of functional decline were apparent in two distinct trajectory groups featuring regular home care use. Furthermore, a small proportion of individuals with a unique set of characteristics was related to the highest levels of ambulatory and emergency care use. New research avenues are outlined to investigate time-varying effects of health factors inside the clusters containing most unfavourable outcomes.

Development of dynamic health care delivery heatmaps for end-of-life cancer care: a cohort study

OBJECTIVE

Measures of variation in end-of-life (EOL) care intensity across hospitals are typically summarised using unidimensional measures. These measures do not capture the full dimensionality of complex clinical care trajectories over time that are needed to inform quality improvement efforts. The objective is to develop a novel visual map of EOL care trajectories that illustrates multidimensional utilisation over time.

SETTING

United States' National Cancer Institute or National Comprehensive Cancer Network (NCI/NCCN)-designated hospitals.

PARTICIPANTS

We identified Medicare claims for fee-for-service beneficiaries with poor prognosis cancers who died between April and December 2016 and received the preponderance of treatment in the last 6 months of life at an NCI/NCCN-designated hospital.

DESIGN

For each beneficiary, we transformed each Medicare claim into two elements to generate a two-dimensional individual-level heatmap. On the y-axis, each claim was classified into a categorical description of the service delivered by a healthcare resource. On the x-axis, the date for each claim was converted into the day number prior to death it occurred on. We then summed up individual-level heatmaps of patients attributed to each hospital to generate two-dimensional hospital-level heatmaps. We used four case studies to illustrate the feasibility of interpreting these heatmaps and to shed light on how they might be used to guide value-based, quality improvement initiatives.

RESULTS

We identified nine distinct EOL care delivery patterns from hospital-level heatmaps based on signal intensity and patterns for inpatient, outpatient and home-based hospice services. We illustrate that in most cases, heatmaps illustrating patterns of multidimensional healthcare utilisation over time provide more information about care trajectories and highlight more heterogeneity than current unidimensional measures.

CONCLUSIONS

This study illustrates the feasibility of representing multidimensional EOL utilisation over time as a heatmap. These heatmaps may provide potentially actionable insights into hospital-level care delivery patterns, and the approach may generalise to other serious illness populations.

Impact of Surgical Wait Time to Hysterectomy for Benign Gynecologic Disease

STUDY OBJECTIVE

To determine the impact of surgical wait time on healthcare use and surgical outcomes for patients undergoing hysterectomy for benign gynecologic indications.

DESIGN

Retrospective cohort study.

SETTING

Urban, academic tertiary care center.

PATIENTS

Patients who underwent hysterectomy for benign disease between 2012 and 2018.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Patients were categorized into 2 groups, dichotomized by surgical wait times >30 days or ≤30 days. Healthcare use was measured by the number of discrete patient interactions with the healthcare system through phone calls, secure electronic messaging, and office and emergency room visits. Univariate and multivariable logistic regression models were performed to assess the association between surgical wait time and healthcare use and perioperative outcomes while controlling for confounders. A total of 277 patients were included in our analysis: 106 (38.3%) had surgical wait times >30 days (median 47 days, range 24-68 days), and 171 (67.1%) had surgical wait times ≤30 days (median 19 days; range 12-26 days). The groups did not differ by age, insurance status, substance use, or comorbid conditions. Patients in the group with surgical wait times >30 days were more likely to have increased healthcare use (69 of 106, 65% vs 43 of 171, 25%; odds ratio 5.55; 95% confidence interval, 3.27-9.41). There were no differences in intraoperative complications (9 of 106, 8% vs 19 of 171, 11%; p = .482) or postoperative complications (28 of 106, 26% vs 32 of 171, 19%; p = .13) between the groups; however, after controlling for potential confounders, patients with surgical wait times >30 days were 3.22 times more likely to be readmitted than patients with surgical wait times ≤30 days (95% confidence interval, 1.27-8.19).

CONCLUSION

A surgical wait time >30 days in patients undergoing a hysterectomy for benign disease is associated with increased healthcare use in the interim. Although patients who experience longer surgical wait times do not experience worse surgical outcomes, they may be at higher risk for readmission after surgery. Targeted interventions to optimize perioperative coordination of care for patients undergoing a hysterectomy for benign disease, especially those within vulnerable populations, are needed to improve quality of care, decrease any redundant or inefficient healthcare use, and reduce any unnecessary delays.

Gaps in Alcohol Screening and Intervention Practices in Surgical Healthcare: A Qualitative Study

OBJECTIVES

Risky alcohol use before surgery is associated with an increased risk of postoperative complications and longer hospital stays. Preoperative alcohol interventions can improve surgical outcomes but are not commonly integrated into routine care. This study sought to better understand patient's and provider's perceptions of alcohol-related surgical health and healthcare practices and illuminate gaps in care and how they could be improved.

METHODS

This study used a descriptive qualitative research design. Data were collected between July 2017 and March 2018. One-on-one interviews assessed domains related to knowledge, gaps in alcohol-related screening and intervention, and interest in enhancing alcohol-related care. Key themes emerged from a process of iterative coding and thematic analysis.

RESULTS

Participants included elective surgical patients who met alcohol screening criteria (n = 20) and surgical healthcare providers (n = 9). Participants had modest or low awareness of alcohol-related surgical health risks. Basic alcohol screening was a routine part of care, but results were often discounted or overlooked. Providers did not routinely initiate preoperative alcohol education or intervention. Providers viewed improving alcohol-related clinical practices as a low priority. Patients were interested in receiving alcohol interventions before surgery if they were delivered in a nonjudgement style and focused on surgical health optimization.

CONCLUSIONS

This study highlights potential gaps in alcohol-related knowledge and care, and found providers place a low priority on alcohol interventions in the perioperative context. Given the high complication rate associated with preoperative alcohol use, these topics are worthy of future research. To be successful strategies to overcome specific barriers to alcohol screening and intervention must address the needs of patients and providers.

Association of Socioeconomic Area Deprivation Index with Hospital Readmissions After Colon and Rectal Surgery

BACKGROUND

Risk adjustment for reimbursement and quality measures omits social risk factors despite adversely affecting health outcomes. Social risk factors are not usually available in electronic health records (EHR) or administrative data. Socioeconomic status can be assessed by using US Census data. Distressed Communities Index (DCI) is based upon zip codes, and the Area Deprivation Index (ADI) provides more granular estimates at the block group level. We examined the association of neighborhood disadvantage using the ADI, DCI, and patient-level insurance status on 30-day readmission risk after colorectal surgery.

METHODS

Our 677 patient cohort was derived from the 2013-2017 National Surgical Quality Improvement Program at a safety net hospital augmented with EHR data to determine insurance status and 30-day readmissions. Patients' home addresses were linked to the ADI and DCI.

RESULTS

Our cohort consisted of 53.9% males and 63.8% Hispanics with a 22.9% 30-day readmission rate from the date of discharge; > 50% lived in highly deprived neighborhoods. Controlling for medical comorbidities and complications, ADI was associated with increased risk of 30 days from the date of discharge readmissions among patients living in medium (OR = 2.15, p = .02) or high (OR = 1.88, p = .03) deprived areas compared to less-deprived neighborhoods, but not insurance status or DCI.

CONCLUSIONS

The ADI identified patients living in deprived communities with increased readmission risk. Our results show that block-group level ADI can potentially be used in risk adjustment, to identify high-risk patients and to design better care pathways that improve health outcomes.

Severe mental illness and health service utilisation for nonpsychiatric medical disorders: A systematic review and meta-analysis

BACKGROUND

Psychiatric comorbidity is known to impact upon use of nonpsychiatric health services. The aim of this systematic review and meta-analysis was to assess the specific impact of severe mental illness (SMI) on the use of inpatient, emergency, and primary care services for nonpsychiatric medical disorders.

METHODS AND FINDINGS

PubMed, Web of Science, PsychINFO, EMBASE, and The Cochrane Library were searched for relevant studies up to October 2018. An updated search was carried out up to the end of February 2020. Studies were included if they assessed the impact of SMI on nonpsychiatric inpatient, emergency, and primary care service use in adults. Study designs eligible for review included observational cohort and case-control studies and randomised controlled trials. Random-effects meta-analyses of the effect of SMI on inpatient admissions, length of hospital stay, 30-day hospital readmission rates, and emergency department use were performed. This review protocol is registered in PROSPERO (CRD42019119516). Seventy-four studies were eligible for review. All were observational cohort or case-control studies carried out in high-income countries. Sample sizes ranged from 27 to 10,777,210. Study quality was assessed using the Newcastle-Ottawa Scale for observational studies. The majority of studies (n = 45) were deemed to be of good quality. Narrative analysis showed that SMI led to increases in use of inpatient, emergency, and primary care services. Meta-analyses revealed that patients with SMI were more likely to be admitted as nonpsychiatric inpatients (pooled odds ratio [OR] = 1.84, 95% confidence interval [CI] 1.21-2.80, p = 0.005, I2 = 100%), had hospital stays that were increased by 0.59 days (pooled standardised mean difference = 0.59 days, 95% CI 0.36-0.83, p < 0.001, I2 = 100%), were more likely to be readmitted to hospital within 30 days (pooled OR = 1.37, 95% CI 1.28-1.47, p < 0.001, I2 = 83%), and were more likely to attend the emergency department (pooled OR = 1.97, 95% CI 1.41-2.76, p < 0.001, I2 = 99%) compared to patients without SMI. Study limitations include considerable heterogeneity across studies, meaning that results of meta-analyses should be interpreted with caution, and the fact that it was not always possible to determine whether service use outcomes definitively excluded mental health treatment.

CONCLUSIONS

In this study, we found that SMI impacts significantly upon the use of nonpsychiatric health services. Illustrating and quantifying this helps to build a case for and guide the delivery of system-wide integration of mental and physical health services.