Sage Journals: Discover world-class research

Abstract

Limited research is available on the real-world experiences of patients with dementia with Lewy bodies (DLB). This study evaluated clinical events, healthcare utilization, and healthcare costs of patients with DLB vs other dementia types with psychosis (ODP). Study patients included commercial and Medicare Advantage with Part D enrollees aged ≥40 years with evidence of DLB and ODP from 6/01/2015‒5/31/2019. Compared with patients with ODP, more patients with DLB had clinical events including anticholinergic effects, neurologic effects, and cognitive decline. Patients with DLB used more healthcare resources with greater dementia-related office and outpatient visits and psychosis-related inpatient stays and office, outpatient, and emergency visits compared with their ODP patient counterparts. Patients with DLB also incurred higher healthcare costs for all-cause and dementia-related office visits and pharmacy fills, and psychosis-related total costs. Understanding the clinical and economic impact of DLB and ODP is important to improve care for patients with dementia.

Keywords

Significance statement

Compared to patients with dementia-related psychosis from other dementia types, patients with dementia with Lewy bodies (DLB) had significantly more anticholinergic and neurologic clinical events, higher all-cause, dementia-related, and psychosis-related healthcare costs, and were more likely to require psychosis-related inpatient stays. These results suggest that clarity about DLB diagnoses are necessary so targeted and safe treatments are evaluated and used to reduce the disease burden in this population.

Introduction

It is estimated that 11% of US adults ages 65 years and older suffer from dementia, a debilitating disorder characterized by memory and personality changes and impaired reasoning.¹ Dementia is the largest single contributor to disability and care needs for older adults out of any chronic condition.² The 2021 costs associated with dementia in the United States were estimated at $355 billion.³ In addition to the high economic burden, studies have also shown a higher burden of concomitant chronic conditions in dementia patients.^4–6 In an analysis of Medicare beneficiaries with dementia, 95% had ≥1 other chronic health condition with 29% having coronary heart disease, 28% congestive heart failure, and 23% diabetes.⁶

Dementia most commonly manifests due to Alzheimer’s disease (AD) but can also take other forms including dementia with Lewy bodies (DLB), vascular dementia, dementia related to Parkinson’s disease, frontotemporal dementia, and mixed dementia. As dementia worsens, neuropsychiatric symptoms, including agitation, euphoria, anxiety, depression, apathy, appetite changes, aberrant motor behavior, delusions, hallucinations, disinhibition, and sleep impairment become more common. An estimated 50%-80% of patients with dementia will experience neuropsychiatric symptoms at some point in the disease course.^7–9 Neuropsychiatric symptoms present a considerable burden to patients and caregivers, including impairments in daily living activities,¹⁰ more rapid cognitive decline,¹¹ worse quality of life,¹² earlier institutionalization,¹³ and greater caregiver depression.¹⁴

Dementia-related psychosis (DRP), which includes delusions and hallucinations, can occur in patients with any dementia type, but is most prevalent in patients with DLB and Parkinson’s disease, with up to 75% experiencing psychosis.¹⁵ This is in comparison to only 30% of patients with AD and 10% of patients with frontotemporal dementia. Due to the lack of viable treatment options for DRP, physicians often treat these patients off-label with antipsychotic medications, despite a box warning advising against antipsychotic use for elderly patients with dementia due to the increased risk of mortality.

After Parkinson’s disease, DLB is the second most commonly diagnosed type of progressive dementia¹⁶ where alpha-synuclein and other proteins develop in nerve cells (ie, Lewy bodies) in the brain regions that control thinking, memory, and movement causing recurrent visual hallucinations, spontaneous extrapyramidal motor issues, rapid eye movement sleep behavior disorder, or cognitive impairment.¹⁷ About 5% of individuals with dementia have only DLB diagnoses and many also have AD pathology.¹⁸ Patients with DLB often have faster disease progression than do patients with other dementia types.^19,20 In Rongve et al, the time to severe dementia was significantly shorter in patients with DLB compared to AD (median 154 day difference).²⁰ Olichney et al, reported that patients with DLB had a faster rate of cognitive decline (P < .001) and a shorter mean survival time from symptoms onset (P = .007) compared with patients with AD.¹⁹ In patients with DLB, first-line treatment for cognitive and psychiatric symptoms should consist of acetylcholinesterase inhibitors (AChEIs), notably rivastigmine.^21,22 Typical antipsychotics are not recommended due to the potential for severe worsening of all symptoms and the development of neuroleptic malignant syndrome (NMS).²³ Benzodiazepines are also not recommended due to the risk of side effects including sedation, increased fall risk, worsening cognition, and paradoxical agitation. If severe behaviors persist after treatment with AChEIs, atypical antipsychotics, notably quetiapine and clozapine, can be used with caution due to their association with severe neuroleptic sensitivity reactions, NMS, worsening parkinsonism, orthostatic hypertension, and drowsiness.^24,25

Limited research is available on the real-world experience of patients with DLB, who constitute a subset of patients with DRP. While costs of care studies are well documented in patients with AD, few studies have looked at patients with DRP due to DLB and other dementia types. Understanding the clinical and economic impact of DRP due to DLB and other dementia types is important for payers and for resource planning from a public health perspective given the rising prevalence of dementia. Additionally, best treatment practices do not exist and there is a lack of information on real-world prescribing patterns for these patients. The objectives of this study were to explore clinical events, healthcare utilization, and healthcare costs for patients with DRP associated with DLB and other dementia types and to understand the unmet needs associated with current management of DLB.

Methods

Data Source

This was a retrospective database study using medical claims, pharmacy claims, and enrollment information from commercial and Medicare Advantage Part D (MAPD) patients during June 01, 2015 to May 31, 2019 (study period) from the Optum Research database (ORD). Medical claims had International Classification of Diseases, ninth and 10^th Revisions, Clinical Modification (ICD-9-CM/ICD-10-CM) diagnosis and procedure codes, Current Procedural Terminology codes, and Healthcare Common Procedure Coding System codes and revenue codes. Outpatient pharmacy codes included National Drug Codes. The study only used de-identified data from the ORD obtained from Covered Entities that permitted de-identification of protected health information (PHI) for use in research studies conducted by Optum. The PHI has been de-identified in accordance with the United States Department of Health and Human Services Privacy Rule’s requirements for de-identification codified at 45 C.F.R. § 164.514(b), and thus, was not subject to the Common Rule requirements and an IRB review.

Study Sample Selection

To be included in the study, patients were required to have one of the following conditions: 1) ≥2 pharmacy fills ≤60 days apart, 2) ≥1 pharmacy fill with ≥45 days supply for a DRP therapy (atypical or typical first-generation antipsychotic, divalproex sodium, or dextromethorphan/quinidine), or 3) ≥1 medical claim with a psychosis diagnosis code (including delusions and hallucinations) in any position and ≥1 pharmacy claim for a DRP therapy ≤365 days apart. The date of the first DRP therapy fill was the index date. Patients were also required to have ≥2 medical claims in any position with a diagnosis of the same type of dementia. The first dementia diagnosis claim was required to have occurred within 180 days of the index date. The second dementia diagnosis claim was required to have occurred within 30-180 days following the first dementia diagnosis. Patients also had continuous health plan enrollment with medical and pharmacy benefits 12 months prior to the index date (pre-index period) and 12 months including and following the index date (post-index period) and were ≥40 years of age during the index year. Patients were excluded if they had a pre-index medical claim with a diagnosis in any position on the claim for a mental health condition manifesting as psychosis; substance use, abuse or dependence associated with psychosis; psychosis due to delirium; cerebral amyloid angiopathy; central nervous system neoplasm, intracranial mass lesion or vascular malformation; ischemic or hemorrhagic stroke; or cortical degeneration. Patients diagnosed with pseudobulbar affect who had a pharmacy claim for dextromethorphan; or diagnosed with epilepsy or seizure who had a pharmacy claim for divalproex sodium during the pre-or post-index periods were also excluded.

Cohorts and Propensity Score Matching (PSM)

Patients were stratified into two cohorts: DLB cohort (patients with ≥2 DLB diagnoses) and other types of dementia-related psychosis (ODP) cohort (patients without ≥2 DLB diagnoses). Patients were 1:1 propensity score-matched based on demographics (age group, sex, insurance type, region); comorbidities (Quan-Charlson comorbidity index score,^26,27 Clinical Classifications Software from the Agency for Healthcare Research and Quality [AHRQ]²⁸); and year of index date. For each DLB cohort patient, an ODP cohort patient with the closest available propensity score was selected, without replacement, within a caliper of ± .01. If there was >1 potential match, the ODP cohort patient with the closest index year was chosen as the match for the DLB cohort patient. Following PSM, bivariate comparisons of demographics and pre-index characteristics were conducted to assess the success of the matching procedure.

Pre-Index Demographic and Clinical Characteristics

Demographic (age, gender, insurance type, geographic region) and clinical characteristics (Charlson comorbidity index; top AHRQ comorbidities; presence of anxiety, depression, and insomnia; dementia type) were measured during the pre-index period.

Study Outcomes

Clinical Events and Mental Health Symptoms

Clinical events were defined as the proportion of patients with ≥1 post-index medical claim with a diagnosis in any position on the claim for extrapyramidal reactions, tardive dyskinesia, neurologic effects (headache, unspecified convulsions, dizziness), anticholinergic effects (dry mouth, blurred vision, constipation), cardiovascular effects, sedation, falls/fractures, or medication-related cognitive effects (≥1 diagnosis code for cognitive decline and ≥1 diagnosis code indicating an adverse effect due to medication use on the same claim). Clinical events were not measured with any temporal relationship to antipsychotic medication consumption.

The presence of mental health symptoms was defined as ≥1 post-index medical claim with a diagnosis in any position on the claim for depressive disorder, anxiety disorder, cognitive decline, or insomnia.

Medication Utilization

Medication utilization was stratified by post-index utilization of antipsychotic medication (pharmacy fill for atypical or typical antipsychotics, divalproex sodium, or dextromethorphan/quinidine) and behavioral health medication (pharmacy fill for selective serotonin reuptake inhibitors [SSRIs], other antidepressants, benzodiazepines, or anti-epileptics/lithium). A line of therapy (LOT) algorithm was used to identify antipsychotic and behavioral health medication regimens. The first LOT included all antipsychotic and behavioral health medications filled within 14 days including and following the index date and ended on the earliest of the following: addition/substitution of a new drug, a gap in treatment of >60 days after the run-out date of all drugs in the LOT, or the end of the study period. Subsequent LOTs were measured if a new regimen was started after the first LOT ended. Subsequent LOTs that occurred after the 12-month post-index period were not observed or included in the analysis.

Healthcare Resource Utilization

All-cause healthcare resource utilization was calculated as the proportion of patients with ≥1 physician office visit, hospital outpatient visit, emergency room visit, or inpatient stay during the pre- and post-index periods. Utilization was considered dementia-related or psychosis-related if the medical claim had a diagnosis code for dementia or psychosis, respectively, in any position on the claim.

Healthcare Costs

All-cause healthcare costs were calculated as the mean combined health plan and patient-paid amounts in the 12-month pre- and post-index periods adjusted for inflation using the annual medical care component of the Consumer Price Index to reflect inflation between the date of the claim and 2020.²⁹ Cost categories included physician office visit, hospital outpatient visit, emergency services visit, inpatient stay, other medical costs, and pharmacy costs. Costs were considered dementia-related or psychosis-related if the claim had a diagnosis code for dementia or psychosis, respectively, in any position on the claim.

Statistical Analyses

Descriptive statistics were calculated for all pre-index and post-index measures. Numbers and percentages were calculated for dichotomous and polychotomous variables. Means and standard deviations were calculated for continuous variables. The balance in demographic and pre-index clinical characteristics between the post-matched DLB and ODP cohorts were evaluated with standardized differences, which show the absolute value of the difference in variable means standardized by the pooled standard deviation.^30,31 Comparisons between DLB and ODP cohorts were tested with Rao-Scott test for binary variables and Wald Z-test using robust standard errors in ordinary least squares regressions for continuous variables. Comparisons between patient-level pre-index vs post-index periods were tested with McNemar test for binary measures and paired t-test for continuous measures.

Results

Pre-Index Demographics and Clinical Characteristics

A total of 442 patients with DLB and 442 matched patients with ODP were included in the study (Figure 1). Patients with DLB were slightly younger than patients with ODP (77.7 years vs 78.8 years, P = .001; Table 1). More than half of patients were male (53.6%), and the majority had MAPD coverage (89.8%). Patients ultimately diagnosed with DLB compared with ODP were more likely to have pre-index dementia diagnoses, which included other dementias (ie, presenile and senile dementia, Pick’s disease, frontotemporal dementia, dementia associated with diseases other than Parkinson’s disease) (59.1% vs 42.5%, P < .001), unspecified dementia (53.4% vs 40.1%, P < .001), AD (25.6% vs 9.7%, P < .001), and Parkinson’s disease with dementia (28.3% vs 2.5%, P < .001). A greater percentage of patients in the DLB cohort had diagnoses for multiple dementia types (55.0% vs 33.0%, P < .001).

Figure 1.

Patient sample selection. CNS, central nervous system; DLB, dementia with Lewy bodies; ODP, other dementia with psychosis.

Table 1.

Pre-index demographic and clinical characteristics.

	Total (n = 884)	DLB Cohort (n = 442)	ODP Cohort (n = 442)	P-value
Age, mean (SD)	78.3 (7.3)	77.7 (7.3)	78.8 (7.3)	.001
Male gender, n (%)	474 (53.6)	230 (52.0)	244 (55.2)	.336
Insurance type, n (%)
Commercial	90 (10.2)	45 (10.2)	45 (10.2)	1.000
MAPD	794 (89.8)	397 (89.8)	397 (89.8)	1.000
Region, n (%)
Northeast	150 (17.0)	76 (17.2)	74 (16.7)	.887
Midwest	238 (26.9)	119 (26.9)	119 (26.9)	1.000
South	385 (43.6)	192 (43.4)	193 (43.7)	.957
West	111 (12.6)	55 (12.4)	56 (12.7)	.900
Charlson comorbidity index, mean (SD)	2.7 (1.7)	2.6 (1.7)	2.8 (1.7)	.122
Top AHRQ comorbidities, n (%)
Delirium	773 (87.4)	385 (87.1)	388 (87.8)	.663
Hypertension	674 (76.2)	337 (76.2)	337 (76.2)	1.000
Other nervous system disorders	633 (71.6)	310 (70.1)	323 (73.1)	.311
Diseases of the urinary system	600 (67.9)	298 (67.4)	302 (68.3)	.646
Disorders of lipid metabolism	557 (63.0)	278 (62.9)	279 (63.1)	.938
Dementia diagnoses (not cohort defining), n (%)
Other dementias^a	449 (50.8)	261 (59.1)	188 (42.5)	<.001
Unspecified dementia	413 (46.7)	236 (53.4)	177 (40.1)	<.001
Multiple dementias	389 (44.0)	243 (55.0)	146 (33.0)	<.001
Alzheimer’s disease	156 (17.7)	113 (25.6)	43 (9.7)	<.001
Parkinson’s Disease with dementia	136 (15.4)	125 (28.3)	11 (2.5)	<.001
Vascular dementia	64 (7.2)	36 (8.1)	28 (6.3)	.314
Dementia with lewy bodies	10 (1.1)	0 (.0)	10 (2.3)	-

^aIncludes presenile and senile dementia, Pick’s disease, frontotemporal dementia, and dementias associated with diseases other than Parkinson’s disease.

AHRQ, Agency for Healthcare Research and Quality; DLB, dementia with Lewy bodies; ODP, other dementia with psychosis.

Post-Index Dementia Diagnoses

In the post-index period, most patients had ≥1 diagnosis for dementia other than the dementia diagnosed closest to the index date (that defined cohorts). A majority had diagnoses for other dementias (69.2%) and unspecified dementia (53.5%). A significantly greater percentage of patients with DLB had ≥1 diagnosis for non-DLB dementias, except for vascular dementia (P < .001 for all comparisons; Table 2). A total of 74.0% of patients with DLB and 48.0% with ODP had post-index diagnoses for multiple dementia types (P < .001).

Table 2.

Post-index diagnoses, clinical events, and symptoms.

	Total (n = 884)	DLB Cohort (n = 442)	ODP Cohort (n = 442)	P-value
Dementia diagnoses (not cohort defining), n (%)
Other dementias^a	612 (69.2)	359 (81.2)	253 (57.2)	<.001
Multiple dementias	539 (61.0)	327 (74.0)	212 (48.0)	<.001
Unspecified dementia	473 (53.5)	271 (61.3)	202 (45.7)	<.001
Alzheimer’s disease	206 (23.3)	135 (30.5)	71 (16.1)	<.001
Parkinson’s Disease with dementia	185 (20.9)	165 (37.3)	20 (4.5)	<.001
Vascular dementia	104 (11.8)	53 (12.0)	51 (11.5)	.835
DLB	15 (1.7)	0 (.0)	15 (3.4)	-
Clinical events, n (%)
Falls and/or fractures	261 (29.5)	122 (27.6)	139 (31.5)	.289
Anticholinergic effects	196 (22.2)	114 (25.8)	82 (18.6)	.031
Neurologic effects	151 (17.1)	84 (19.0)	67 (15.2)	.040
Cardiovascular effects	43 (4.9)	20 (4.5)	23 (5.2)	.595
Sedation	21 (2.4)	15 (3.4)	6 (1.4)	.063
Extrapyramidal reactions	13 (1.5)	6 (1.4)	7 (1.6)	.793
Tardive dyskinesia	5 (.6)	3 (.7)	2 (.5)	.581
Cognitive effects due to medication	4 (.5)	3 (.7)	1 (.2)	.343
Behavioral and mental health disorders, n (%)
Depressive disorders	463 (52.4)	234 (52.9)	229 (51.8)	.668
Anxiety disorders	370 (41.9)	184 (41.6)	186 (42.1)	.883
Cognitive decline	181 (20.5)	109 (24.7)	72 (16.3)	<.001
Insomnia disorders	167 (18.9)	94 (21.3)	73 (16.5)	.059

^aIncludes presenile and senile dementia, Pick’s disease, frontotemporal dementia, and dementias associated with diseases other than Parkinson’s disease.

DLB, dementia with Lewy bodies; ODP, other dementia with psychosis.

Clinical Events and Mental Health Symptoms

Almost 30% of patients had a fall and/or fracture and more than 20% experienced anticholinergic effects during the post-index period. Patients in the DLB and ODP cohorts did not differ in most measures of clinical events (Table 2); however, patients with DLB were more likely to experience anticholinergic effects (25.8% vs 18.6%, P = .031) and neurologic effects (19.0% vs 15.2%, P = .040) compared with patients with ODP.

Mental health diagnoses were more common than clinical events. More than half of all study patients (52.4%) had a diagnosis for depression disorders and 41.9% had a diagnosis for anxiety disorders (Table 2). Patients in the DLB cohort were more likely to experience post-index cognitive decline than patients in the ODP cohort (24.7% vs 16.3%, P < .001).

Medication Utilization

Atypical antipsychotics were prescribed to 92.0% of study patients during the post-index period, most commonly quetiapine (68.1%), risperidone (18.3%) and olanzapine (12.4%); data not shown). A greater percentage of patients with DLB were prescribed an atypical antipsychotic during the first LOT (58.6% vs 43.4%, P < .001), but the percentages were not significantly different across the second and third LOTs (Figure 2). Use of atypical antipsychotics alone decreased between 82%‒85% from the first to the second LOT. Among the top nine treatment regimens (prescribed to 86% of patients), there were no significant differences between cohorts in the length of time on first-line therapy. SSRIs and other antidepressants were the most common adjunctive treatments in the second LOT, particularly among patients with DLB. Patients in the DLB cohort were less likely to be prescribed divalproex sodium as first-line treatment compared to patients with ODP (2.3% vs 6.3%, P = .009). As their second regimen, patients in the DLB cohort were more likely to be prescribed a single medication (26.5% vs 21.2%, P = .034) while patients in the ODP cohort were more likely to be prescribed combination therapy (73.5% vs 78.8%, P = .034). Third-line treatment patterns did not differ significantly between the cohorts.

Figure 2.

Most common treatment regimens by line of therapy. Bias may have been introduced from breaking the match when comparing patients with and without a subsequent line of therapy after the first line of therapy. *P < .05 in comparison of DLB and ODP cohorts. APs, antipsychotics; DLB, dementia with Lewy bodies; LOT, line of therapy; ODP, other dementia with psychosis; SSRIs, selective serotonin reuptake inhibitors.

Healthcare Resource Utilization

Almost all patients from both cohorts had a pre-index all-cause office and outpatient visit (Supplementary Figure 1). Patients with DLB were significantly more likely to have a pre-index office visit compared with patients with ODP (94.8% vs 87.6%, P < .001). More than half of patients in both cohorts had a pre-index all-cause emergency visit and approximately one-third had an inpatient stay. Patients with DLB were less likely to have a pre-index all-cause inpatient stay compared with patients with ODP (27.4% vs 36.9%, P = .001).

During the post-index period, almost all patients had an all-cause office and outpatient visit, more than 60% had an emergency visit, and more than one-third had an inpatient stay (Figure 3). Post-index all-cause healthcare utilization was similar between the cohorts, except patients with DLB were significantly more likely to have an office visit compared with patients with ODP (90.7% vs 82.1%, P < .001).

Figure 3.

Mean post-index healthcare resource utilization. *P < .05 in comparison of DLB and ODP cohorts. DLB, dementia with Lewy bodies; ODP, other dementia with psychosis.

All-cause healthcare utilization was comprised mostly of dementia-related care (Figure 3). Compared to patients with ODP, those with DLB were significantly more likely to have a dementia-related office visit (79.2% vs 61.3%, P < .001) and outpatient visit (72.0% vs 60.4%, P = .014). Patients with DLB also had higher psychosis-related healthcare utilization than patients with ODP (Figure 3). The percentage of patients with a psychosis-related office or outpatient visit was more than double in patients with DLB vs ODP (office: 27.4% vs 11.3%, P < .001; outpatient: 17.0% vs 5.7%, P < .001). The percentage of patients with a psychosis-related emergency visit or inpatient stay was more than 40% higher in patients with DLB vs ODP (emergency: 11.1% vs 7.7%, P = .014; inpatient: 14.0% vs 9.1%, P = .020).

In comparisons of the pre-index vs post-index period, all-cause healthcare utilization was not significantly different (Supplementary Figure 1), except that all-cause office visits were significantly higher in the pre-index vs post-index period and the proportion of patients with DLB with an all-cause inpatient stay was significantly higher in the post-index vs pre-index period (37.8% vs 27.4%, P < .001).

Healthcare Costs

Pre-index mean all-cause healthcare costs were driven by inpatient stays and pharmacy costs, representing approximately 33% and 25% of the total costs among patients with DLB and 42% and 24% among patients with ODP (Supplementary Figure 2). While total pre-index mean all-cause costs were similar between the two cohorts, patients with DLB had significantly higher mean all-cause office visit costs compared with patients with ODP ($1,956 vs $1,285, P < .001).

Like pre-index costs, post-index mean all-cause costs were also driven by inpatient and pharmacy costs (Figure 4). Patients with DLB had significantly higher post-index all-cause costs for office visits ($1,497 vs $1,065, P < .001) and pharmacy fills ($6,616 vs $4,515, P = .003) compared with patients with ODP.

Figure 4.

Mean post-index healthcare costs. *P < .05 in comparison of DLB and ODP cohorts. DLB, dementia with Lewy bodies; ODP, other dementia with psychosis.

Dementia-related costs were responsible for more than half of the total post-index all-cause costs ($11,754 of $21,111) (Figure 4). Costs for inpatient stays comprised approximately 61% of the total dementia-related costs ($7,149 of $11,754). Patients with DLB had significantly higher mean dementia-related costs for pharmacy ($810 vs $675, P = .049) and office visits ($507 vs $301, P < .001). Patients with DLB also had significantly higher mean costs for post-index psychosis-related care compared with patients with ODP ($5,801 vs $2,687, P < .001) (Figure 4). Patients with DLB had significantly higher mean psychosis-related costs for pharmacy fills, inpatient stays, office visits, and other medical expenses. Psychosis-related costs were driven heavily by inpatients stays and pharmacy costs, which accounted for 51.2% and 40.4%, respectively, of total psychosis-related costs in patients with DLB and 68.9% and 23.2%, respectively in patients with ODP. Patients with DLB had pharmacy costs 276% higher ($2,343 vs $623, P < .001) and inpatient costs 61% higher ($2,971 vs $1,851, P = .004) compared with patients with ODP.

In pre-index vs post-index comparisons, mean all-cause total healthcare costs were significantly higher in the post-index period compared to the pre-index period across both cohorts (Supplementary Figure 2). Both cohorts had significantly higher mean pharmacy, emergency, outpatient, and other medical costs in the post-index period. Patients with DLB also accrued significantly higher mean inpatient costs during the post-index period compared to the pre-index period ($8,046 vs $5,392, P = .008). Conversely, both cohorts saw significantly higher mean office visit costs in the pre-index vs post-index period (DLB: $1,956 vs $1,497, P < .001; ODP: $1,285 vs $1,065, P = .003).

Discussion

This study documented the clinical and economic burden among patients with DLB compared with patients with ODP. Compared to patients with ODP, patients with DLB were more likely to have diagnoses of other dementia, Parkinson’s disease dementia, AD, unspecified dementia, and diagnoses for multiple dementia types. It is possible that patients with DLB often receive diagnoses of other dementia types given the difficulty in accurately diagnosing these patients. The 1-year rule distinguishing between DLB and Parkinson’s disease may be difficult to apply in clinical settings.³² Patients with DLB were also more likely to have anticholinergic and neurologic symptoms during the post-index period. Patients with DLB were over 50% more likely to have evidence of cognitive decline in the post-index period compared with patients with ODP. While there are no studies directly comparing patients with DLB to those with ODP, this finding is like those from previously published research that found patients with DLB often reported worse outcomes in the rate of cognitive decline compared to patients with other dementia types.^19,20,33 In one study, patients with DLB had a more rapid annual rate of decline than patients with AD or Parkinson’s disease dementia.³³ Another study reported that a DLB diagnosis was associated with a shorter time to severe dementia or death in comparison with patients with AD (HR = 2.04, P = .02).²⁰

While both cohorts included patients with evidence of psychosis, this study suggests that patients with DLB had more severe psychosis symptoms that led to medical treatment. Compared to patients with ODP, patients with DLB had consistently higher utilization of office visits; dementia-related outpatient visits; and psychosis-related outpatient visits, emergency visits, and inpatient stays. While the proportion of patients with DLB with an office visit decreased in the post-index period, the proportion of patients requiring an inpatient stay increased significantly following their DLB diagnosis. Again, there is a lack of previously published studies comparing DLB to ODP in regards to health economic outcomes; however, in comparisons with AD, patients with DLB had higher risks for hospital admission.^34–36 In Mueller et al, the rate of hospital admission within the year after a dementia diagnosis was significantly higher in patients with DLB vs AD (IRR = 1.4, P < .001).³⁵ This was true for both planned and unplanned hospital admissions, as well as the mean number of hospital days. The length of stay in the Mueller study was determined by neuropsychiatric symptoms. As both cohorts in our study had psychosis, it may explain why we did not see a significant difference in the proportion of patients with claims for an all-cause hospital stay between patients with DLB and ODP. However, we did find that patients with DLB were significantly more likely to require a psychosis-related inpatient stay. In Hanyu et al, patients with DLB had approximately double the risk for hospital admission or death compared with patients with AD (30% vs 14%, P < .005).³⁴ Additionally, a US-based study of patients with DLB showed hospital stays to be 7 ± 8 days on average, with worsening confusion or hallucinations as the most common reason for hospitalization.³⁶

Patients with DLB had significantly higher mean all-cause, dementia-related, and psychosis-related costs for pharmacy fills and office visits, as well as mean psychosis-related inpatient costs. Studies investigating the costs of DLB compared to other dementia types are limited. In one study, patients with DLB had an additional $4,055 in annual medical costs compared with patients with AD ($12,081 vs $8,027); however, this difference was not significant.³⁷ Contrastingly, compared to patients with AD, a Swedish study reported significantly higher all-cause costs³⁸ and a UK study found significantly higher hospitalization costs among patients with DLB³⁵ driven by poor physical health and neuropsychiatric symptoms. In the present study, patients with DLB had an excess of $1,120 in mean annual psychosis-related inpatient costs compared to patients with ODP.

A significantly higher percentage of patients with DLB (58.6% vs 43.4%) were prescribed atypical antipsychotics as first-line therapy. Olanzapine and risperidone were the second and third most commonly prescribed atypical antipsychotics despite warnings against use due to higher incidence of serious side effects. Atypical antipsychotics have a known association with severe neuroleptic sensitivity reactions, NMS, worsening parkinsonism, orthostatic hypotension, and drowsiness among patients with DLB.^1,24,25 It is possible that the higher number of different dementia diagnoses before and after their first fill for a DRP medication and the lack of clarity regarding their dementia type may have contributed to potentially unsafe treatment for nearly 60% of patients with DLB. The higher incidence of clinical events among patients with DLB also suggests that potentially unsafe treatment could explain higher utilization and costs for office visits and psychosis-related inpatient stays.

Patients with DRP face considerable clinical and economic burdens³⁹ and patients with DLB fare even worse, with more clinical events and greater healthcare utilization and costs. This study raises awareness of the high economic and clinical burden of patients with DLB and the unmet need for medications to effectively control psychiatric symptoms. Future longitudinal studies are needed to assess the differences in the disease course and management of patients with psychosis due to DLB and other dementia types.

Limitations

Certain limitations exist in claims database analyses and results should be interpreted with that in mind. The presence of a diagnosis code on a medical claim is not proof of disease, as it may be incorrectly coded or included as rule-out criteria rather than actual disease. To mitigate this, >1 pharmacy fill or a combination of a diagnosis and pharmacy fill was required to identify patients. The index date was designated based on the first observed antipsychotic pharmacy fill, but not necessarily when a physician reached a diagnosis. Cohorts were defined by provider diagnoses as confirmation of laboratory or imaging tests were not available in claims data. Clinical events were not measured with any temporal relationship to antipsychotic consumption, and thus should not be interpreted as resulting directly from antipsychotic therapy. The utilization of dementia medications, specifically rivastigmine, which is recommended as first-line treatment for DLB, was not tracked in this study and should be a consideration in future research on medication utilization. This study had a fixed 12-month post-index period and not every patient had a second or third LOT, which necessitated a breaking of the match. As a result, this may have introduced bias in the P-values for subsequent LOTs. The second- and third-line treatment results should be viewed as a description of all those that occurred within 12 months. Treatments prescribed after the 12-month post-index period were not included in this analysis. Additionally, depression, anxiety, and insomnia disorders were not incident measures and patients may have been diagnosed with these conditions prior to the start of the study. Twenty-four months of continuous health plan enrollment was required, which may have introduced selection bias. Patients who disenrolled (including those who died) prior to meeting enrollment requirements were excluded, potentially resulting in healthier patients retained for the study. Claims data do not contain clinical information, such as the duration, severity, or frequency of dementia-related symptoms. Lastly, this study was conducted in a large US managed care population and may not be generalizable to other populations or all patients with a DRP diagnosis.

Conclusions

This study documented the high rates of clinical events, healthcare resource utilization, and healthcare costs among patients with DLB and ODP. Compared to patients with ODP, patients with DLB diagnoses had significantly more anticholinergic and neurologic clinical events, higher all-cause and dementia-related costs for pharmacy fills and office visits, higher psychosis-related costs, and were more likely to require psychosis-related inpatient stays. The higher all-cause and dementia-related costs for pharmacy fills suggest that clarity about DLB diagnoses are necessary so targeted and safe treatments are evaluated and used to reduce the disease burden in this population.

Supplemental Material

Supplemental Material - Burden of Illness Among Patients with Psychosis due to Dementia with Lewy Bodies and Other Dementias

Supplemental Material for Burden of Illness Among Patients with Psychosis due to Dementia with Lewy Bodies and Other Dementias by Monica Frazer, PhD, Nazia Rashid, PharmD, MS, Scott Bunner, MPH, Ben Skoog, PharmD, and Victor Abler, MD in American Journal of Alzheimer's Disease & Other DementiasÂ®

Supplemental Material

Supplemental Material - Burden of Illness Among Patients with Psychosis due to Dementia with Lewy Bodies and Other Dementias

Footnotes

Acknowledgments

This study was funded by Acadia Pharmaceuticals. Deja Scott-Shemon, MPH, an employee of Optum, provided medical writing assistance. Sharanya Murali, an employee of Optum, provided project management assistance.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: MF, SB, and RH were employees of Optum at the time the study was conducted and were funded by Acadia Pharmaceuticals to conduct the study. NR, BS, and VA were employees of Acadia Pharmaceuticals at the time the study was conducted.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Acadia Pharmaceuticals, who participated in the study design, interpretation of study results, and critical review of the manuscript.

ORCID iDs

Monica Frazer

Nazia Rashid

Supplemental Material

Supplemental material for this article is available online.

Appendix

References

Hudomiet

Hurd

Rohwedder

. Dementia prevalence in the United States in 2000 and 2012: Estimates based on a nationally representative study. J Gerontol B Psychol Sci Soc Sci. 2018;73:S10-S9.

Wimo

APM

. World Alzheimer Report 2010: The Global Economic Impact of Dementia. London: Alzheimer's Disease International; 2010.

Hurd

Martorell

Delavande

Mullen

Langa

. Monetary costs of dementia in the United States. N Engl J Med. 2013;368:1326-1334.

Hill

Futterman

Duttagupta

Mastey

Lloyd

Fillit

. Alzheimer’s disease and related dementias increase costs of comorbidities in managed Medicare. Neurology. 2002;58:62-70.

Bynum

JPW

Rabins

Weller

Niefeld

Anderson

. The relationship between a dementia diagnosis, chronic illness, medicare expenditures, and hospital use. J Am Geriatr Soc. 2004;52:187-194.

Association

. Alzheimer's Disease and Chronic Health Conditions: The Real Challenge for 21st Century Medicine; 2020. http://www.alz.org/national/documents/report_chroniccare.pdf Accessed 9 12, 2020.

Rabins

PVLC

Steele

. Practical Dementia Care. New York, NY: Oxford University Press; 1999.

Lyketsos

Lopez

Jones

Fitzpatrick

Breitner

DeKosky

. Prevalence of neuropsychiatric symptoms in dementia and mild cognitive impairment: Results from the cardiovascular health study. JAMA. 2002;288:1475-1483.

Lyketsos

Steinberg

Tschanz

Norton

Steffens

Breitner

. Mental and behavioral disturbances in dementia: findings from the Cache County Study on Memory in Aging. Am J Psychiatr. 2000;157:708-714.

10.

Lyketsos

Steele

Baker

, et al. Major and minor depression in Alzheimer’s disease: Prevalence and impact. J Neuropsychiatry Clin Neurosci. 1997;9:556-561.

11.

Stern

Tang

Albert

, et al. Predicting time to nursing home care and death in individuals with Alzheimer disease. JAMA. 1997;277:806-812.

12.

Gonzalez-Salvador

Lyketsos

BAker

, et al. Quality of life in dementia patients in long-term care. Int J Geriatr Psychiatry. 2000;15:181-189.

13.

Steele

Rovner

Chase

Folstein

. Psychiatric symptoms and nursing home placement of patients with Alzheimer’s disease. Am J Psychiatr. 1990;147:1049-1051.

14.

Gonzalez-Salvador

Arango

Lyketsos

Barba

. The stress and psychological morbidity of the Alzheimer patient caregiver. Int J Geriatr Psychiatry. 1999;14:701-710.

15.

Cummings

Ballard

Tariot

, et al. Pimavanserin: Potential treatment for dementia-related psychosis. J Prev Alzheimers Dis. 2018;5:253-258.

16.

Heidebrink

. Is dementia with Lewy bodies the second most common cause of dementia? J Geriatr Psychiatry Neurol. 2002;15:182-187.

17.

Mckeith

. Dementia with lewy bodies: A clinical overview. In: Ames

O'Brien Burns

, eds Dementia. Boca Raton: CRC Press; 2017.

18.

2020 Alzheimer’s disease facts and figures. Alzheimers Dement. 2020;16(3):391-460.

19.

Olichney

JMGD

Galasko

Salmon

, et al. Cognitive decline is faster in Lewy body variant than in Alzheimer’s disease. Neurology. 1998;51:351-357.

20.

Rongve

Soennesyn

Skogseth

, et al. Cognitive decline in dementia with Lewy bodies: A 5-year prospective cohort study. BMJ Open. 2016;6:e010357.

21.

McKeith

Del Ser

Spano

, et al. Efficacy of rivastigmine in dementia with Lewy bodies: A randomised, double-blind, placebo-controlled international study. Lancet. 2000;356:2031-2036.

22.

Rozzini

Chilovi

Bertoletti

, et al. Cognitive and psychopathologic response to rivastigmine in dementia with Lewy bodies compared to Alzheimer’s disease: A case control study. Am J Alzheimers Dis Other Demen. 2007;22:42-47.

23.

McKeith

FAirbAirn

PeRRy

ThomPson

Perry

FAirbAirn

. Neuroleptic sensitivity in patients with senile dementia of Lewy body type. BMJ 1992;305:673-678.

24.

Walker

Grace

Overshot

, et al. Olanzapine in dementia with Lewy bodies: A clinical study. Int J Geriatr Psychiatry. 1999;14:459-466.

25.

Culo

Mulsant

Rosen

, et al. Treating neuropsychiatric symptoms in dementia with Lewy bodies: A randomized controlled-trial. Alzheimer Dis Assoc Disord. 2010;24:360-364.

26.

Bayliss

Ellis

Shoup

Zeng

McQuillan

Steiner

. Association of patient-centered outcomes with patient-reported and ICD-9-based morbidity measures. Ann Fam Med. 2012;10:126-133.

27.

Quan

Couris

, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173:676-682.

28.

HCUP CCS [computer Program]. Rockville, MD: Agency for Healthcare Research and Quality; 2015. Available at: https://www.hcup-us.ahrq.gov/toolssoftware/ccs/ccs.jsp

29.

US Department of Labor, Bureau of Labor Statistics . Consumer Price Index. Medical Care. Series ID: CUUR0000SAM. Washington, DC: US Department of Labor, Bureau of Labor Statistics; 2020. http://data.bls.gov/cgi-bin/surveymost?cu Accessed 9 12, 2020.

30.

Austin

. Assessing balance in measured baseline covariates when using many-to-one matching on the propensity-score. Pharmacoepidemiol Drug Saf. 2008;17:1218-1225.

31.

Austin

. An introduction to propensity score methods for reducing the effects of confounding in observational studies. Multivariate Behav Res. 2011;46:399-424.

32.

McKeith

Boeve

Dickson

, et al. Diagnosis and management of dementia with Lewy bodies: Fourth consensus report of the DLB consortium. Neurology. 2017;89:88-100.

33.

Kramberger

Auestad

Garcia-Ptacek

, et al. Long-term cognitive decline in dementia with lewy bodies in a large multicenter, international cohort. J Alzheimers Dis. 2017;57:787-795.

34.

Hanyu

HST

SaTo

Hirao

Kanetaka

Sakurai

IwamoTo

. Differences in clinical course between dementia with Lewy bodies and Alzheimer’s disease. Eur J Neurol. 2009;16:212-217.

35.

Mueller

Perera

Rajkumar

, et al. Hospitalization in people with dementia with Lewy bodies: Frequency, duration, and cost implications. Alzheimers Dement (Amst). 2018;10:143-152.

36.

Spears

Besharat

Monari

Martinez-Ramirez

Almeida

Armstrong

. Causes and outcomes of hospitalization in Lewy body dementia: A retrospective cohort study. Parkinsonism Relat Disord. 2019;64:106-111.

37.

Zhu

Scarmeas

Stavitsky

, et al. Comparison of costs of care between patients with Alzheimer’s disease and dementia with Lewy bodies. Alzheimers Dement. 2008;4:280-284.

38.

Bostrom

FJL

Jönsson

Minthon

Londos

. Patients with Lewy body dementia use more resources than those with Alzheimer’s disease. Int J Geriatr Psychiatry 2007;22:713-719.

39.

Frazer

Abler

Halpern

Skoog

Rashid

. Burden of illness among patients with dementia-related psychosis. J Manag Care Spec Pharm. 2021;27:367-378.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB

0.08 MB