Emergency department visits in Parkinson’s disease: The impact of comorbid conditions





Abstract


Introduction


Older adults have complex, often overlapping, medical conditions requiring careful management that may lead to increased emergency department usage compared to younger adults. Parkinson’s disease (PD), a progressive neurodegenerative disorder characterized by distinct motor and nonmotor features, frequently occurs with additional comorbid disease. Classifying comorbid conditions into clinical subgroups allows for further understanding of the heterogeneity in outcomes in patients with PD. The current study examines the reasons for emergency department (ED) visits in a cohort of patients with PD and identifies comorbidities that are potential risk factors for specific ED presenting conditions.


Methods


Using data from Optum’s de-identified Integrated Claims-Clinical dataset years 2010–2018, patients with PD were identified based on ICD-9/10 diagnosis codes. We identified all ED visits occurring after the first observed diagnosis code for PD. Comorbid conditions were classified using the AHRQ Clinical Classification Software (CCS). We classified patients using Latent Class Analysis (LCA) and conducted multiple logistic regression models with the outcome of reason-for-visit to examine the associations with comorbidity-profile class, patient demographics, and socio-economic characteristics.


Results


The most common reasons for ED admission were injuries such as fractures and contusions, diseases of the circulatory system, and general signs and symptoms, including abdominal pain, malaise, and fatigue. Comorbid medical conditions often observed in this patient population include depression, diabetes mellitus, and chronic pulmonary disease. Patients in the “Poorest Health” classification of the LCA had greater odds for ED admission for diseases related to the gastrointestinal system, musculoskeletal system, and injury/poisoning categories and reduced odds for admission for diseases of the circulatory system.


Discussion


Patients with PD who present to the emergency department with injuries are more likely to be in poor health overall with a high comorbidity burden. Clarifying the complex medical needs of patients with PD is the first step to further individualize care, which may reduce ED visits in this population, improve quality of life, and lessen the footprint on the healthcare system.


Highlights




  • •

    Presenting to the emergency department with injuries may indicate poor health overall.


  • •

    Patients requiring emergency department care tend to have a high comorbidity burden.


  • •

    Comorbidity profiles correlated with primary reason for emergency department visit.


  • •

    Reasons for visit could be used to target prevention and intervention in this group.



1


Introduction


In the general United States population, increased age is associated with more emergency department (ED) visits [ ]. Evidence shows ED visits vary by sex, race, ethnicity, insurance type, and age [ ]. Over the last 20 years the percentage of patients over the age of 60 having one or more ED visits per year has increased from 22% to 26.8% [ ], suggesting ED usage is rising among older adults. Adverse health outcomes following ED visits in the older adult population include functional decline, hospitalization, and ED readmission [ ]. Parkinson’s disease (PD) is a progressive neurodegenerative disorder characterized by distinct motor and nonmotor features with an estimated prevalence of 930,000 individuals in North America in 2020 [ ]. The age of onset for PD is typically 60–70 years old [ , ]. Approximately one-third of patients with PD are seen in an ED or hospitalized each year, and increased age and longer disease duration are associated with more hospitalizations [ ]. Patients with PD are vulnerable to adverse outcomes following ED visits due to the risks associated with being older adults in addition to the severity and complexity of their disease.


PD patients have a wide range of clinical needs which may complicate ED presentation. ED presentations may occur due to direct complications of PD such as worsening of movement disorders, adverse drug effects or neuropsychiatric manifestations; indirect complications such as falls or infections; complications of deep brain stimulators; and unrelated emergencies [ , ]. PD patients often present to the ED for falls/fractures [ ], infections [ ], cardiovascular emergencies [ ], cerebrovascular emergencies [ ], altered mental status [ ], and abdominal pain [ , ]. While prior work has examined the primary reasons for presentation to the ED among patients with PD, most do not consider the clinical complexity due to comorbid conditions in this population. PD patients have reported an average of 7 ± 2 comorbidities [ ] across most organ systems with more than half of reported comorbidities being gastrointestinal and include symptoms such as drooling, difficulty swallowing, delayed gastric emptying, constipation, nausea, and bloating [ ]. Depression, anxiety, and sleep disorders can also be clinically significant in PD patients [ ], with depression present in 35–40% of patients with PD [ , ]. Cardiovascular comorbidities include cardiac autonomic dysfunction, cardiomyopathy, coronary artery disease, and arrhythmias [ ] and may present as postural instability or cognitive decline in PD patients [ ]. Orthostatic hypotension in PD patients increases the risk of falls requiring ED visit and hospitalization [ ]. Evidence also suggests an increased risk of myocardial infarction and stroke following PD diagnosis [ ]. It is unclear if or how these comorbid conditions influence ED utilization within patients with PD. Additionally, many prior works examining ED use within patients with PD have been limited to single institution or small health system studies [ ]. Thus, we have little understanding of the relationship between PD, comorbid conditions, and ED utilization. In the present study, we identified a cohort of patients with PD in a nationally distributed electronic health record (EHR) database to examine the relationship between comorbid conditions and reason for ED visit within a population of patients with PD with the goal of providing information to physicians to predict what may prompt an ED visit in patients based on their current comorbidities.


2


Methods


2.1


Data source


This study examined EHR data from the Optum’s de-identified Integrated Claims-Clinical dataset between January 1, 2008 and December 31, 2018. Data included records of outpatient, ED, and inpatient medical services; diagnostic codes; procedure codes; laboratory values; and prescribed and administered medication from a national integrated healthcare data delivery network that captured data from major healthcare systems and organizations in 50 states and the District of Columbia. Encounter settings included academic, non-academic, and safety-net points of care. Records of institutionalized patients or from military healthcare facilities were excluded. Source data was a sample of 5 million adult patients (≥ 18 years) throughout the United States who were commercially insured, government insured, and uninsured.


2.2


Study design and participant sample


This analysis includes all patients in the source dataset with a diagnosis of primary Parkinson’s Disorder (ICD9 332.0 “paralysis agitans” and ICD10 G.20 “Parkinson’s Disorder”) with at least one visit to the ED at any time in the available integrated dataset following the index PD diagnosis ( N = 3165). The index diagnosis of PD was defined as the first inpatient diagnosis or first outpatient diagnosis, provided a second outpatient diagnosis followed the first.


2.3


Measures


The primary outcome measure in this study was the reason for ED visit. Reason for visit was determined by the primary diagnosis for each emergency visit and classified using the Agency for Healthcare Research and Quality (AHRQ) Clinical Classification Software (CCS) [ ]. For patients with multiple ED visits (Table S1), analysis was limited to the first ED visit following the index PD diagnosis.


The explanatory variables of interest included comorbid conditions, demographic information, and socioeconomic markers. Comorbid conditions were assessed using diagnosis codes recorded during any healthcare encounter (outpatient and inpatient) for the 6 months prior to the patient’s ED visit (Table S2). Comorbidities were categorized using the Charlson-Deyo Comorbidity Index [ ] with the addition of depression, anxiety, bipolar disorder, and schizophrenia [ ].


All demographic and socioeconomic variables were examined at the time of index PD diagnosis. Age was determined from reported birth year, demographic variables were obtained from the EHR, and socioeconomic variables (region, average household income per ZIP code, and proportion of individuals with college education) were extracted from United States Census data. Gender, race, and ethnicity were self-reported. Patients were classified with a single insurer type from the following: commercial insurance, Medicare, Medicaid, dual eligible Medicare and Medicaid, other, and uninsured/unknown.


2.4


Latent class analysis


We conducted a Latent Class Analysis (LCA) to classify patients based on pre-existing comorbid conditions in the 6 months prior to the first ED visit following the index PD diagnosis. Comorbid conditions were categorized as present or absent. We tested two, three, four, and five class solutions and selected the model of best performance in fit statistics (AIC and entropy; Table S3). The character of each class was assigned based on the comorbidity profile within the latent class. Latent class served as the primary predictor variable for subsequent logistic regressions.


2.5


Statistical analysis


Two-tailed student’s t -tests and Wilcoxon U tests were used to analyze continuous demographic variables including age and household income across two groups: PD patients with ED visits and PD patients without ED visits. Chi-square tests were used to compare data from categorical variables such as race, region, and insurance status. The 10 most frequent CCS classifications for reason for ED visit were selected as outcomes for logistic regression. Unadjusted and adjusted logistic regression models for each outcome were calculated using comorbid condition latent class as the primary predictor. Adjusted models controlled for patient demographic and socioeconomic characteristics including age, gender, race, and primary payor. All the statistical tests were two tailed and statistical significance examined at p = 0.05. The statistical analyses were performed in SAS 9.4 (SAS Institute, Cary, NC, USA).


3


Results


Of the 7487 eligible PD patients, 42% documented at least one ED visit ( Table 1 ). PD patients with an ED visit were similar to PD patients without ED visits on most socio-demographic measures. The geographic distribution of patients with PD in this dataset is primarily in the Northeastern and Southern United States in contrast to other studies where PD is more common in the Midwest and Northeastern US [ ]. Individuals with an ED visit had a greater proportion of Medicare coverage than those without an ED visit ( Table 1 ).



Table 1

Sample demographics.







































































































































































































All PD patients Without ED visits With ED visits p -value
( n = 7487) ( n = 4322) ( n = 3165)
Age (years)
Mean (SD) 73.7 (10.5) 73.5 (10.3) 74.1 (10.8) 0.010
Median (IQR) 75 (68–82) 75 (67–82) 76 (68–83) <0.001
Gender (%) 0.068
Female 43.9 43.1 45.0
Male 56.1 56.9 55.0
Unknown 0.0 0.0 0.1
Race/Ethnicity (%) <0.001
Non-Hispanic White 86.7 86.3 87.4
Non-Hispanic African American 4.8 4.2 5.5
Hispanic 2.9 2.8 3.1
Other/unknown 5.6 2.8 4.1
Average household income per ZIP, US dollar (thousands)
Mean (SD) 42.6 (10.5) 43.0 (11.4) 41.9 (9.1) <0.001
Median (IQR) 40.1 (35.7–47.7) 40.3 (35.7–48.2) 40.1 (35.7–46.9) 0.122
College educated residents per ZIP (%)
Mean (SD) 24.3 (8.0) 24.5 (8.3) 24.0 (7.4) 0.007
Median (IQR) 23 (18–28) 23 (18–29) 23 (18–27) 0.280
Region (%) <0.001
Northeast 48.9 46.1 52.6
Midwest 12.7 14.4 10.5
Other/unknown 2.9 2.9 2.9
South 26.2 27.7 24.3
West 9.3 9.0 9.7
Primary payor (%) <0.001
Any Commercial 44.7 46.3 42.5
Medicare/Medicaid Dual Eligible 3.9 3.6 4.3
Medicare only 38.4 37.0 40.3
Medicaid only 1.5 1.2 2.0
Other 0.8 0.8 0.8
Unknown/Uninsured 10.7 11.2 10.1


The most common reason for ED visits in this cohort was injury and poisoning, representing 20.94% of ED admissions ( Table 2 ). This category of diagnoses includes injuries due to fractures, sprains, and contusions as well as adverse effects related to ingested chemicals or improper use of medications [ ]. Diseases of the circulatory system were the primary reason for visit in 13.17% of cases ( Table 2 ) and describe admissions due to myocardial infarction, heart failure, and hypertension with complications, among others [ ]. 11.12% of ED visits were attributed to general signs and symptoms ( Table 2 ), which include abdominal pain, malaise, and fatigue [ ].



Table 2

Reasons for visit.








































































Reasons for ED visits in PD patients Frequency Percent
Injury and poisoning 4609 20.94
Diseases of the circulatory system 2900 13.17
Symptoms; signs; and ill-defined conditions and factors influencing health status 2448 11.12
Diseases of the respiratory system 2134 9.69
Diseases of the musculoskeletal system and connective tissue 1927 8.75
Diseases of the nervous system and sense organs 1801 8.18
Residual codes & Unclassified 1539 6.99
Diseases of the genitourinary system 1322 6.01
Diseases of the gastrointestinal system 1153 5.24
Mental Illness 783 3.56
Endocrine; nutritional; and metabolic diseases and immunity disorders 484 2.2
Diseases of the skin and subcutaneous tissue 443 2.01
Infectious and parasitic diseases 348 1.58
Diseases of the blood and blood-forming organs 69 0.31
Neoplasms 46 0.21
Congenital anomalies 7 0.03


Comorbid medical conditions in this population were categorized using the Charlson-Deyo Comorbidity Index [ ] with the addition of depression, anxiety, bipolar disorder, and schizophrenia [ ]. The most frequently observed comorbidity in this cohort was depression, seen in 34.61% of patients. PD patients visiting the ED also often had diabetes mellitus at a rate of 30.53% and chronic pulmonary disease in 30.37% of cases ( Table 3 ).


Tags:
Mar 29, 2024 | Posted by in EMERGENCY MEDICINE | Comments Off on Emergency department visits in Parkinson’s disease: The impact of comorbid conditions

Full access? Get Clinical Tree
Get Clinical Tree app for offline access