Epidemiology and economic burden of Huntington’s disease: a Canadian provincial public health system perspective

Abstract

Aims

To evaluate the epidemiology, healthcare resource utilization, and direct healthcare costs associated with Huntington’s disease in a Canadian setting with a universal healthcare system.

Materials and methods

Using Albertan administrative health data, a retrospective cohort was identified applying an algorithm requiring two HD diagnostic codes within two years, using the first record as the index date (i.e. proxy for diagnosis date), from 1 April 2010 to 31 March 2019 for patients ≥21 years old. Incidence/prevalence measures were evaluated from 1 April 2010 to 31 March 2019, while healthcare resource utilization and healthcare costs per person-year (inflated to 2020 Canadian dollars) were evaluated from index to the end of follow-up (death, moved out of province, or 31 March 2020).

Results

Mean [standard deviation] age at index (n = 395) was 53.9 [13.8] years and 53.7% were female. From 2010 to 2019, annual HD incidence varied between 0.47 and 1.21/100,000 person-years and HD prevalence increased from 7.25 to 9.33/100,000 persons. The mean number of visits per person-year for general and specialist practitioners was 19.2 [18.8] and 12.2 [25.5], respectively. The mean total all-cause direct healthcare costs were $23,211 [$38,599] per person-year, with hospitalizations accounting for 57.8% of all-cause costs. Costs were higher among individuals with a long-term care stay, a proxy for disease severity.

Limitations and conclusions

This study utilizes administrative health data to describe the epidemiology of HD and utilization of publicly funded care by individuals with HD. While administrative data presents limitations since it is not collected for research purposes, it provides a population-level examination of the burden of HD. There was a substantial economic burden associated with HD in a Canadian setting.

Keywords:

• Huntington’s disease
• epidemiology
• burden of illness
• healthcare resource utilization
• chronic disease
• neurodegenerative diseases

JEL CLASSIFICATION CODES:

• I00
• I
• I12
• I1
• I

Background

Huntington’s disease (HD) is a rare neurological disorder with autosomal dominant inheritance, with typical onset between 30 and 50 years of age^1,2. HD has an estimated disease prevalence of 4.0–13.9 individuals per 100,000 in the general population in Canada and up to 17.2 per 100,000 in the Caucasian population specifically, using data primarily collected in the province of British Columbia^3–5. Estimates extrapolated from Fisher and Hayden³ suggest that there are up to 4,700 individuals affected with HD and 14,000 at 50% risk for HD in Canada. Current evidence suggests that the prevalence of HD may be increasing, both in Canada and globally^3,5, which may be due to a combination of factors including: improved disease-specific knowledge, greater availability of genetic testing, diminished stigma around HD diagnosis⁵, expanding cytosine, adenine, and guanine repeats over time⁶, increasing mutation rates, increasing population life expectancy, and greater availability of symptomatic HD treatments (e.g. anti-choreiform, antidepressant medications)⁵.

Without currently available treatments to prevent HD or slow its progression, the management of HD requires a multidisciplinary approach focused on maximizing the function and quality of life for individuals diagnosed with the disease¹. The progressive cognitive, motor, and psychiatric impairment associated with HD can increase the care needs of affected individuals, increasing the risk of hospitalization and earlier placement into long-term care^7,8. Consequently, the burden of HD on patients, families, and the healthcare system is significant and increases with disease progression.

Direct healthcare costs for HD have been estimated in the United States (US) to be between $3,257 and 4,947 US Dollars (USD; 2011 values) in the early stages, rising steeply to $22,582–37,495 USD in the late stages (annualized cost per patient)⁸. In the United Kingdom (UK), mean annual costs per person with Stage I HD were £2,250, rising sharply to £89,760 (2013 UK pounds: $141,164 in 2011 USD^9,10) per person with Stage IV HD¹¹. Despite this burden, there is limited evidence evaluating the healthcare resource utilization and costs of HD in Canada⁸, and cost estimates available from other countries may not be applicable due to differences between the healthcare systems^7,8,11–14. Therefore, a real-world evidence study examining the epidemiology, healthcare resource utilization, and direct healthcare costs from the Canadian public healthcare perspective was conducted.

Methods

Study design and population

We conducted a retrospective observational study of a cohort of individuals with HD in Alberta, Canada using province-wide administrative health data from 1 April 2010 to 31 March 2020. Deidentified data were released by Alberta Health (Government of Alberta Ministry of Health) for the following linkable administrative datasets: Alberta Blue Cross Pharmacy Claims, Alberta Continuing Care Information System, Alberta Precision Laboratories Dataset, Diagnostic Imaging Dataset, Discharge Abstract Database, National Ambulatory Care Reporting System, Pharmaceutical Information Network Dispenses, Population Registry, Practitioner Claims, and Vital Statistics (deaths). Data from the Practitioner Claims, Discharge Abstract Database, and National Ambulatory Care Reporting System datasets were requested starting from 1 April 1997 to facilitate identification of incident and prevalent cases.

Cases were identified from 1 April 2010 until 31 March 2019 based on a published algorithm¹³, requiring two diagnostic codes within a two-year span from the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM)/International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, Canada (ICD-10-CA), appearing in any position in the Discharge Abstract Database or National Ambulatory Care Reporting System (Supplementary Table 1). Inclusion in the cohort was determined by the earliest of the two diagnostic codes (index date, used as a proxy for the diagnosis date), meeting the definition algorithm for HD. Patients whose index dates were within the case ascertainment period were included as incident cases. Patients who met the algorithm criteria prior to 1 April 2010 were considered prevalent cases and 1 April 2010 was used as their index date. Juvenile HD was excluded from the study population using the age at index date (i.e. <21 years old^15,16). To examine healthcare resource utilization and associated costs, a burden of illness study cohort was established, requiring at least one year of follow-up data.

Study variables

To describe the study cohort, demographic characteristics including age, sex, and Alberta Health Services geographic zone were extracted from the Population Registry at index date. Additionally, clinical characteristics of patients were derived, including a modified Charlson Comorbidity Index score (one-year prior to index date)¹⁷ and huntingtin (HTT) genetic test status (captured at any time based on data from Alberta Precision Laboratories). Long-term care stay during follow-up was used as a proxy for disease severity. Annual incidence and prevalence rates were calculated for each of the 2010/11–2018/19 fiscal years, while the five-year average incidence and period prevalence rates were calculated for the most recent fiscal years (2014/15–2018/19). Healthcare resource utilization and healthcare costs were estimated (details in Supplementary Table 2) and examined across the follow-up period per person-year, from index date to the end of follow-up (date of death, date moved out of the province, or 31 March 2020). All-cause, HD-related, and HD-specific healthcare resource utilization and costs were calculated for the following: hospitalizations, physician visits (stratified into general practitioners and specialist practitioners), ambulatory care (including emergency department and day surgery) visits, and length of stay in hospital. HD-related healthcare resource utilization was defined as a HD diagnostic code in any position, while HD-specific healthcare resource utilization was defined as a HD diagnostic code included as the primary reason for a visit. Only HD-related prescription medications (i.e. vesicular monoamine transporter 2 inhibitors, antipsychotics, antidepressants, and benzodiazepines) were extracted and these cost estimates were used for the all-cause and HD-specific medication cost estimates. Only HD-related diagnostic imaging records were extracted (head computerized tomography and brain magnetic resonance imaging). Long-term care stays and diagnostic imaging were considered all-cause as diagnostic code information is not included with these healthcare resource utilization components; therefore, the all-cause cost for long-term care and diagnostic imaging was also included in the HD-related and HD-specific cost calculations.

Statistical analysis

Patient demographic and clinical characteristics, healthcare resource utilization, and costs were analyzed per person year of follow-up using descriptive statistics, including frequencies, proportions, means, medians, standard deviations (SDs), and interquartile ranges (IQRs). Annual incidence rates were calculated as the number of newly identified patients with HD each year divided by the estimated mid-year Alberta adult population (≥21 years old) in each respective year. The annual incidence rates are presented per 100,000 person-years. Annual prevalence rates were calculated as the total number of patients with HD divided by the same population denominators and presented per 100,000 persons. The five-year average annual incidence was calculated as a weighted average of the annual incidence rates for the most recent period (2014/15–2018/19). The five-year period prevalence was calculated as the number of patients with HD identified by the algorithm in Alberta between 1 April 2014 and 31 March 2019 divided by the mid-year Alberta adult (≥21 years of age) population of the middle year (2016/17).

Healthcare resource utilization and cost outcomes were calculated per person-year (i.e. per person per year) of follow-up. Patients who did not use a particular resource were assigned a value of zero for the healthcare resource utilization and cost. The Statistics Canada Consumer Price Index annual inflation percentage change was used to adjust all costs to 2020 Canadian dollars¹⁸.

All analyses were performed using SAS 9.4 (SAS Institute, Cary, North Carolina). This study was approved by the Health Research Ethics Board of Alberta—Community Health Committee.

Results

Overall, 418 individuals met the algorithm criteria and of these, 395 individuals had ≥1 year of follow-up data (Supplementary Figure 1), with a mean [SD] duration of follow-up of 6.2 [2.9] years (data not shown). The mean age at index was 53.9 [13.8] years and over half (53.7%) of patients were female (Supplementary Table 3).

The annual HD incidence rates varied between 0.47 and 1.21 per 100,000 person-years, and the five-year average annual HD incidence was 0.83 per 100,000 person-years (Table 1). The prevalence of HD increased from 7.25 to 9.33 per 100,000 persons between 2010/11 and 2018/19, with a five-year period prevalence of 12.15 per 100,000 persons.

Table 1. Annual (2010–2018) and average 5-year (2014–2018) incidence and prevalence of HD in Alberta.

Fiscal year	Number of incident patients	Person-years	Incidence (per 100,000 person-years)	Number of prevalent patients	Mid-year population	Prevalence (per 100,000)
2010/11	26	2,745,527	0.95	199	2,745,527	7.25
2011/12	13	2,793,009	0.47	212	2,793,009	7.59
2012/13	26	2,862,175	0.91	230	2,862,175	8.04
2013/14	28	2,947,228	0.95	251	2,947,228	8.52
2014/15	26	3,027,608	0.86	265	3,027,608	8.75
2015/16	37	3,070,416	1.21	283	3,070,416	9.22
2016/17	22	3,104,520	0.71	289	3,104,520	9.31
2017/18	23	3,139,143	0.73	295	3,139,143	9.40
2018/19	21	3,183,874	0.66	297	3,183,874	9.33
5-year (2014/15–2018/19)	129	15,525,561	0.83	368	3,027,608	12.15

Abbreviation. HD, Huntington’s disease.

Note: Annual incidence and prevalence estimates were calculated using the estimated mid-year adult (≥21 years of age) population of Alberta in each respective year; the 5-year incidence is a weighted average of the 5 most recent annual incidence estimates; the 5-year period prevalence was estimated using the mid-year population of the middle year (2016).

Overall, 236 (59.7%) individuals were hospitalized at least once during the follow-up period. While there were low numbers of hospitalizations per person-year for the entire burden of illness cohort (Table 2), the average number of cumulative days spent in hospital per person-year of follow-up was just under two weeks (all-cause: 13.9 [36.3] days; HD-related and HD-specific: 12.0 [35.2] days). Among all hospitalizations, the mean number of days per hospitalization was longer for HD-specific hospitalizations (165.1 [361.1] days vs. all-cause: 57.0 [192.7] days and HD-related: 74.9 [226.9] days; Supplementary Figure 2).

Table 2. Healthcare resource utilization among patients with HD per person-year of follow-up.

Treatment/health services classification HD patients (n = 395)	Number per person-year
	Mean (SD)	Median (IQR)
Prescription medication dispenses^a^,b
Patients with ≥1 prescription dispense, n (%)	300 (75.9)
Number of dispenses	19.9 (33.7)	5.3 (0.1–21.5)
Hospital visits^a
All-cause	0.3 (0.4)	0.1 (0.0–0.4)
Total days in hospital^c	13.9 (36.3)	1.0 (0.0–10.7)
Patients with ≥1 hospitalization, n (%)	236 (59.7)
HD-related	0.2 (0.3)	0.0 (0.0–0.3)
Total days in hospital^c	12.0 (35.2)	0.0 (0.0–7.6)
HD-specific	0.1 (0.1)	0.0 (0.0–0.1)
Total days in hospital^c	12.0 (35.2)	0.0 (0.0–7.6)
Physician visits—general practitioner^a
All-cause	19.2 (18.8)	11.1 (4.8–29.6)
HD-related	5.0 (10.1)	0.9 (0.2–4.2)
HD-specific	4.7 (10.0)	0.8 (0.2–3.9)
Physician visits—specialist practitioner^a
All-cause	12.2 (25.5)	5.1 (2.4–11.6)
HD-related	1.0 (2.7)	0.5 (0.0–1.1)
HD-specific	0.9 (2.4)	0.5 (0.0–1.0)
Ambulatory care visits^a
All-cause	4.9 (9.8)	2.9 (1.3–5.5)
HD-related	1.1 (2.7)	0.5 (0.0–1.2)
HD-specific	0.8 (2.4)	0.1 (0.0–0.8)
Long-term care stays
Number of stays^a	0.1 (0.1)	0.0 (0.0–0.1)
Length of stay (days)^a	7.7 (48.2)	0.0 (0.0–0.0)
Patients with ≥1 long-term care stay, n (%)	149 (37.7)
Number of stays	0.2 (0.2)	0.2 (0.1–0.3)
Length of stay (days)	20.5 (77.0)	0.0 (0.0–0.4)
Long-term care stays—convalescent stays^d
Number of stays^a	0.0 (0.1)	0.0 (0.0–0.0)
Length of stay (days)^a	0.1 (0.4)	0.0 (0.0–0.0)
Patients with ≥1 convalescent stay, n (%)	12 (3.0)
Number of stays	0.2 (0.2)	0.2 (0.1–0.3)
Length of stay (days)	1.9 (1.5)	1.4 (0.9–2.6)
Diagnostic imaging tests^a^,e
CT (head)	0.1 (0.3)	0.0 (0.0–0.2)
MRI (brain)	0.0 (0.1)	0.0 (0.0–0.0)

Abbreviation. CT, computerized tomography; HD, Huntington’s disease; HRU, healthcare resource utilization; ICU, intensive care unit; IQR, interquartile range; MRI, magnetic resonance imaging; SD, standard deviation.

^aHealthcare resource utilization was averaged across all patients over the follow-up period. Patients who had no record of a particular service were assigned a value of zero for that service.

^bPrescription medications included: vesicular monoamine transporter 2 [VMAT2] inhibitors, antipsychotics, antidepressants, benzodiazepines, donepezil hydrochloride, levodopa/carbidopa, amantadine hydrochloride, galantamin hydrobromide, memantine hydrochloride, rivastigmine.

^cDays in hospital was calculated as a sum of days spent in hospital (across multiple hospitalizations per patient, if applicable) during follow-up and averaged across all patients with HD per person-year, where those who were not hospitalized were assigned a value of zero.

^dConvalescent stays were defined as stays ≤2 weeks.

^eDiagnostic imaging tests were obtained from the Diagnostic Imaging dataset.

The mean all-cause costs per person-year was $23,211 [$38,599], while the HD-related and HD-specific costs were $16,755 [$34,195] and $12,960 [$29,465], respectively (Figure 1(A)). Hospitalizations were the largest cost driver, with a mean per person-year of $13,422 [$33,972] for all-cause (57.8%), $11,175 [$32,657] (66.7%) for HD-related, and $7,584 [$28,072] (58.5%) for HD-specific hospitalizations (Supplementary Table 4).

Figure 1. Total costs (mean [SD]) in CAD for the (A) overall HD cohort (n = 395) and (B) stratified by long-term care stay, per person-year of follow-up. CAD: Canadian dollars; HD: Huntington’s disease; SD: standard deviation. Notes: All costs reflect 2020 CAD and were adjusted for inflation using the Statistics Canada Consumer Price Index¹⁸. Costs were averaged across all patients, where patients who did not have a particular healthcare resource utilization were assigned a value of zero. Costs for prescriptions dispensed, diagnostic imaging, and long-term care could not be separated into HD-related and HD-specific and were held constant across categories.

The mean number of all-cause, HD-related, and HD-specific physician visits per person-year of follow-up were higher for general practitioners compared to specialist practitioners (Table 2). However, the costs for these visits were lower for general practitioners compared to specialist practitioners for all-cause visits ($1,218 [$1,259] vs. $1,948 [$3,629]), but higher for HD-related ($291 [$584] vs. $155 [$399]) and HD-specific visits ($262 [$568] vs. $131 [$290], Supplementary Table 4).

To evaluate the impact of disease severity on healthcare resource utilization and costs, results were stratified by the presence of a long-term care stay (37.7% with a long-term care stay; Table 2). Individuals with a long-term care stay had a higher proportion of hospitalizations (77.9%; not specifically during the long-term care stay) compared to those without (48.8%), and a longer mean all-cause length of stay (26.1 [43.5] days) compared to those without (6.6 [28.9] days; Table 3). Individuals with a long-term care stay had a more than 3-fold higher mean number of all-cause general practitioner visits (35.0 [18.0] vs. 9.6 [11.2]), and a more than 2-fold higher mean number of all-cause specialist practitioner visits (18.3 [34.1] vs. 8.5 [17.4]), compared to those without. Similar patterns were observed for both HD-related and HD-specific hospitalizations, and general practitioner and specialist practitioner visits. The increased healthcare resource utilization among individuals with a long-term care stay equated to markedly higher mean all-cause costs ($41,370 [$38,853]) compared to those without ($12,213 [$34,089]; Figure 1(B)). This was particularly evident across hospitalizations (Supplementary Table 5), with the mean cost for those with a long-term care stay being higher compared to those without a long-term care stay (all-cause: $23,259 [$35,607] vs. $7,463 [$31,554]; HD-related: $20,174 [$34,331] vs. $5,725 [$30,398]; HD-specific: $13,435 [$26,644] vs. $4,039 [$28,372], respectively).

Table 3. Healthcare resource utilization among patients with HD per person-year of follow-up stratified by long-term care stay.

Treatment/health services classification	Number per person-year
	No long-term care stay (n = 246)		With long-term care stay (n = 149)
	Mean (SD)	Median (IQR)	Mean (SD)	Median (IQR)
Prescription medication dispenses^a^,b
Patients with ≥1 prescription dispense, n (%)	183 (74.4)		117 (78.5)
Cost of dispenses	10.3 (19.0)	3.5 (0.0–12.5)	35.9 (44.8)	12.8 (0.4–62.7)
Hospital visits^a
All-cause	0.2 (0.4)	0.0 (0.0–0.3)	0.4 (0.5)	0.3 (0.1–0.5)
Total days in hospital^c	6.6 (28.9)	0.0 (0.0–2.3)	26.1 (43.5)	9.8 (0.8–30.2)
Patients with ≥1 hospitalization, n (%)	120 (48.8)		116 (77.9)
HD-related	0.1 (0.3)	0.0 (0.0–0.1)	0.3 (0.3)	0.2 (0.0–0.4)
Total days in hospital^c	5.2 (28.3)	0.0 (0.0–0.9)	23.1 (42.1)	6.8 (0.0–25.6)
HD-specific	0.0 (0.1)	0.0 (0.0–0.0)	0.1 (0.2)	0.0 (0.0–0.2)
Total days in hospital^c	5.2 (28.3)	0.0 (0.0–0.9)	23.1 (42.1)	6.8 (0.0–25.6)
Physician visits—general practitioner^a
All-cause	9.6 (11.2)	6.0 (3.3–10.4)	35.0 (18.0)	34.8 (20.2–46.8)
HD-related	1.8 (4.7)	0.5 (0.0–1.4)	10.4 (13.7)	4.9 (1.0–13.4)
HD-specific	1.6 (4.6)	0.4 (0.0–1.1)	9.9 (13.6)	4.5 (1.0–11.8)
Physician visits—specialist practitioner^a
All-cause	8.5 (17.4)	4.7 (2.4–8.8)	18.3 (34.1)	6.4 (2.7–16.5)
HD-related	0.8 (0.9)	0.6 (0.1–1.1)	1.2 (4.2)	0.4 (0.0–0.9)
HD-specific	0.8 (0.9)	0.6 (0.1–1.1)	1.0 (3.7)	0.4 (0.0–0.9)
Ambulatory care visits^a
All-cause	5.7 (11.6)	3.3 (1.6–6.2)	3.5 (5.8)	2.2 (0.9–3.8)
HD-related	1.0 (1.6)	0.4 (0.0–1.3)	1.3 (3.9)	0.5 (0.1–1.1)
HD-specific	0.7 (1.2)	0.1 (0.0–0.9)	0.8 (3.5)	0.1 (0.0–0.7)
Long-term care stays
Number of stays^a	–	–	0.2 (0.2)	0.2 (0.1–0.3)
Length of stay (days)^a	–	–	20.5 (77.0)	0.0 (0.0–0.4)
Patients with ≥1 long-term care stay, n (%)	0 (0.0)		149 (100.0)
Number of stays	–	–	0.2 (0.2)	0.2 (0.1–0.3)
Length of stay (days)	–	–	20.5 (77.0)	0.0 (0.0–0.4)
Long-term care stays—convalescent stays^d
Number of stays^a	–	–	0.0 (0.1)	0.0 (0.0–0.0)
Length of stay (days)^a	–	–	0.1 (0.6)	0.0 (0.0–0.0)
Patients with ≥1 convalescent stay, n (%)	0 (0.0)		12 (8.1)
Number of stays	–	–	0.2 (0.2)	0.2 (0.1–0.3)
Length of stay (days)	–	–	1.9 (1.5)	1.4 (0.9–2.6)
Diagnostic imaging tests^a^,e
CT (head)	0.1 (0.2)	0.0 (0.0–0.1)	0.2 (0.4)	0.0 (0.0–0.3)
MRI (brain)	0.0 (0.1)	0.0 (0.0–0.0)	0.0 (0.1)	0.0 (0.0–0.0)

Abbreviation. CT, computerized tomography; HD, Huntington’s disease; HRU, healthcare resource utilization; ICU, intensive care unit; IQR, interquartile range; MRI, magnetic resonance imaging; SD, standard deviation; VMAT2, vesicular monoamine transporter 2.

^aHealthcare resource utilization was averaged across all patients over the follow-up period. Patients who had no record of a particular service were assigned a value of zero for that service.

^bPrescription medications included: vesicular monoamine transporter 2 [VMAT2] inhibitors, antipsychotics, antidepressants, benzodiazepines, donepezil hydrochloride, levodopa/carbidopa, amantadine hydrochloride, galantamin hydrobromide, memantine hydrochloride, rivastigmine.

^cDays in hospital was calculated as a sum of days spent in hospital (across multiple hospitalizations per patient, if applicable) during follow-up and averaged across all patients with HD per person-year, where those who were not hospitalized were assigned a value of zero.

^dConvalescent stays were defined as stays ≤2 weeks.

^eDiagnostic imaging tests were obtained from the Diagnostic Imaging dataset.

Discussion

The present study found that the prevalence of HD is increasing in Alberta. Substantial healthcare resource utilization and costs associated with HD were also found, primarily driven by physician visits and hospitalizations, respectively, with both healthcare resource utilization and costs being higher among individuals with a long-term care stay.

Estimated annual HD incidence varied between 0.47 and 1.21 per 100,000 person-years in the overall cohort (n = 418), which is generally consistent with estimates from the international literature of retrospective real-world studies that ranged from 0.30 to 1.59 per 100,000 person-years in Italy^16,19, Spain²⁰, and the US²¹. The only Canadian study of HD incidence was from 2001, and while there were considerable differences in study design, they reported a similar annual incidence rate to this study of 6.9 per million (0.69 per 100,000) person-years²².

The prevalence of HD increased over the study period (from 7.25 to 9.33 per 100,000 persons) in the overall cohort, similar to trends reported in other studies^13,23,24 and from a systematic review⁵. Results from our study fall between the prevalence estimates from two studies in British Columbia (4.0–13.9 per 100,000)^3,4 and align with overall Canadian survey estimates⁴. Outside of Canada, global reviews have also shown relatively similar estimates in North America and Europe^6,25. Overall, incidence and prevalence data support the validity of the approach adopted in our study to define an HD cohort, strengthening the results herein reported on the economic burden of HD in a Canadian setting.

We identified a burden of illness cohort of 395 individuals with HD (incident and prevalent patients with HD with at least one year of follow-up), with most between 50 and 59 years of age at index, which is consistent with other results found in the literature^{13,15,16,19,21,23}.

The results from our study indicate greater healthcare resource utilization among individuals with HD compared to the Canadian average. The mean number of general practitioner (19.2 [18.8]) and specialist practitioner (12.2 [25.5]) visits per person-year observed in our study population was higher than the estimated Canadian average physician visits (6.7) in 2019²⁶. This difference may have been the result of visits during the high rates of long-term care admission (37.7%), indicating that patients had substantial comorbidities or were in poor health. Similarly, the mean number of all-cause hospital admissions per person-year (0.3 [0.4]) in our study was higher than the reported 2018 average (0.08) for Canada²⁶. The median HD-specific length of stay in the current study was 65 days per hospitalization, which is notably higher than reports for related neurological conditions in Canada, such as Alzheimer’s disease (21 days), multiple sclerosis (7 days) and Parkinson’s disease (14 days)²⁷. All-cause length of stay was also higher in this study than reported by the Institute for Health Economics in 2014–2015 (10 vs. 7.5 days)²⁸. Further research is necessary to understand the drivers of the longer length of stay observed, however, contributing factors could be the variety of cognitive, motor, and psychiatric impairments associated with HD progression, and discharge from hospital being dependent on a patient securing and accepting space in a long-term care facility.

This study provides the first Canadian estimates of direct healthcare costs in HD. In 2021, the forecasted average healthcare expenditure per person (public only) in Canada and Alberta was $5,987 and $6,037, respectively²⁹, which is substantially lower than the costs per patient-year observed in our study (all-cause: $23,211; HD-related: $16,755; HD-specific: $12,960), but differing methodology in the study design, costing components, and time period reflected makes direct comparison challenging. In the 2010/11 fiscal year, using administrative data, the British Columbia Ministry of Health estimated direct healthcare costs per capita were 5.8 times higher for HD, 6.0 times higher for Parkinsonism, 3.0 times higher for Alzheimer’s disease and other dementias, and 3.7 times higher for multiple sclerosis compared to age-standardized data for individuals without these conditions³⁰. Similarly, although we cannot directly compare these results due to methodological differences, the direct healthcare cost estimates provided by our study provide some quantification for the substantial costs associated with HD in a Canadian setting.

Hospitalizations were the largest driver of costs in our study cohort with HD-related hospital costs notably representing 83% of all-cause hospital costs. As was observed with healthcare resource utilization, the mean all-cause ($13,422 [$33,972]), HD-related ($11,175 [$32,657]), and HD-specific ($7,584 [$28,072]) hospitalization costs for our HD cohort were higher compared to the average cost of a standard hospital stay in Canada in 2015 ($5,789)²⁸. The longer length of stay in hospital, particularly for HD-specific and HD-related hospitalizations, indicates a substantial hospitalization burden associated with HD, which may contribute to hospitalizations being the largest cost driver in the present study.

In general, results of the present study are in line with existing literature on healthcare resource utilization and cost associated with HD in other jurisdictions. Divino et al.⁸ found mean total annualized costs for HD ranging from $4,947 to $22,582 per patient for patients with private insurance and $3,257 to $37,495 for patients receiving Medicaid in the US (2011 USD). With respect to the distribution of cost related to HD care, the present study found that medication and diagnostic testing represent only a small portion of the total cost. A similar finding was made by Jones et al.¹¹. There was substantial heterogeneity in healthcare resource utilization and costs associated with long-term care reported in the literature, reflective of heterogeneity in care models between and within jurisdictions, with previous studies reporting longer lengths of stay and higher costs, depending on insurance status, for recent studies conducted in the US^7,8. This study observed high rates of general practitioner and specialist practitioner visits per person-year of follow-up, and associated costs are higher compared to estimates reported in other countries including the US, France, Italy, Germany, and Poland^31,32. However, the estimates are not directly comparable, and additionally, Dorey et al.³¹ cautioned that due to differences in patient severity at the time of diagnosis and care practices, comparisons between countries are difficult to interpret. The higher all-cause physician costs observed in our study may be related to the higher general practitioner and specialist practitioner costs associated with long-term care stays.

To our knowledge, this is the first study to evaluate the burden of disease, healthcare resource utilization, and direct costs associated with HD in a Canadian setting. The use of multiple province-wide datasets allowed for a comprehensive understanding of the healthcare resource utilization and associated costs from the provincial public healthcare system perspective among a cohort of individuals with HD. Results of this study indicate a substantial burden of disease with the primary cost driver being hospitalizations, also in addition to a high number of physician visits, and over one-third of the population having at least one long-term care stay during follow-up.

Some limitations should be noted when interpreting these results. Administrative health data are collected for the purpose of healthcare administration rather than research, which may impact the information collected. ICD-9-CM/ICD-10-CA codes are not a confirmed presence of disease, as the diagnosis could have been included to rule out other conditions or coded incorrectly. The fluctuations observed in the annual incidence and prevalence estimates in our study may be a result of the algorithm used to identify patients for inclusion. For incident cases, the first date of the appearance of an HD diagnosis code within two years was used as the index date and for prevalent cases, the index date was the beginning of the case ascertainment period (1 April 2010). However, the index date used for this study may not truly be reflective of an individual’s clinical diagnosis date. In addition, the absolute number of patients in our study cohort, particularly for annual incidence, diagnosed each year was very small and may contribute to estimate fluctuations. While we reported on those who had the HTT genetic test performed by Alberta Precisions Laboratories, most patients did not have results available (either not tested or tested in another province or service); therefore, molecular confirmation of HD was not included as an inclusion criterion for our study population. The generalizability of the HD patient population identified from this study may be limited, as this study only captured patients with healthcare system interactions due to their HD, thus the algorithm used may preferentially identify patients with later stage HD. Lastly, the healthcare resource utilization and associated costs estimated in this study only captured direct healthcare costs from the perspective of the provincial public healthcare system in Alberta, since indirect cost information (e.g. care partner burden, lost wages and benefits, absenteeism, presenteeism) was not available in the administrative health data in Alberta. The burden of illness from the societal perspective is likely much larger as this would include costs paid by private insurance or patients’ out-of-pocket costs.

Conclusion

Results of this study show that the prevalence of HD is increasing in Alberta and that HD is associated with substantial healthcare resource utilization and costs, driven primarily by physician visits and hospitalizations. Higher healthcare utilization and costs were also associated with the presence of a long-term care stay, which was used as a proxy for disease severity. Gaining this understanding of the prevalence and burden of HD is critical given the pattern of increasing healthcare resource utilization with increasing disease severity. This information highlights potential opportunities to improve care for individuals with HD in Canadian settings, namely, in community-based strategies that may reduce the number and length of hospitalizations, the main cost driver in our study.

Transparency

Declaration of funding

This work was supported by Hoffmann-La Roche Ltd, Mississauga, Ontario, Canada.

Declaration of financial/other relationships

This study is based on data provided by Alberta Health and Alberta Precision Laboratories. The interpretation and conclusions contained herein are those of the researchers and do not necessarily represent the views of the Government of Alberta. Neither the Government of Alberta nor Alberta Health expresses any opinion in relation to this study.

ES, MM, PE, SM, EG, and TC are employed by Medlior Health Outcomes Research Ltd. which received funding for the study from Hoffmann-La Roche Ltd.

JWW, NB, and BM are employed by Hoffmann-La Roche Ltd., who funded this study.

JWW, NB, and BM also hold Hoffmann-La Roche Ltd. stock.

TAM reports speaker honorarium from Abbvie, and International Parkinson and Movement Disorder Society; consultancies from CHDI Foundation/Management, Sunovion, Valeo Pharma, Roche, nQ Medical and Merz; advisory board from Abbvie; Biogen, Sunovion, Medtronic, and research funding from EU Joint Programme—Neurodegenerative Disease Research, uOBMRI, Roche, Ontario Research Fund, CIHR, MJFF, Parkinson Canada, PDF/PSG, LesLois Foundation, PSI Foundation, Parkinson Research Consortium and Brain Canada.

A reviewer on this manuscript has disclosed they have had an educational consultancy for PTC paid to their institution—not on HD. They are also on the advisory boards for Novartis (HD), and received MEGS awards from Sanofi, Akcea and AMGEN for academic work and genetic testing not related to HD. Their centre has received reimbursement for clinical trials from Roche and Prilenia, and research grant funding from the Scottish Government, University of Aberdeen development trust, MRC and HTA. They have attended meetings sponsored by Akcea, MSD, AMGEN, Sanofi, Novartis, Teva and Daiichi Sankyo. The other peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Author contributions

ES, MM, PE, SM, EG, TC, JWW, NB, BM, and TAM were involved in the conception and design of the study, the analysis and interpretation of the data, as well as drafting the paper, revising it critically for intellectual content and approving the final version of the paper to be published. All authors agree to be accountable for all aspects of the work.

Previous presentations

A small portion of these results were presented in a poster at the virtual 28th Annual Meeting of the Huntington Study Group (HSG) from 4 to 6 November 2021³³.

Acknowledgements

No assistance in the preparation of this article is to be declared.