Medical cost and healthcare utilization of amyotrophic lateral sclerosis in China

Abstract Amyotrophic lateral sclerosis (ALS), a specific neurodegenerative disease, imposed increased economic and utilizations burden on the healthcare system, especially with the progress of the diseases severity. However, the economic burden on Chinese ALS patients remained unclear. This study therefore was aimed to investigate medical cost and healthcare utilization for Chinese ALS patients. Longitudinal health data of over 20 million individuals, including military personnel and civilians, was collected from all Chinese military hospitals. We identified 480 patients with a first major diagnosis for ALS from 2015 to 2018, while matched 400 controlled patients on age, gender, ethnic group, geographic region, length of stay, year of diagnosis and comorbidity. Their medical cost and healthcare utilizations were then measured 1 year before, and 1 year after ALS diagnosis. The median annual medical cost of ALS patients was about 2-fold higher, 17,087 CNY during the index year than 1 year before, 7859 CNY. The highest increase in utilizations may account for medical costs on ALS patients, which was represented by hospitalizations (Odd Ratio (OR) = 4.26, 95% confidence interval (CI) 3.52, 5.15), electromyography (OR = 4.14, 95% CI 2.37, 7.22), nerve conduction velocity (OR = 3.26, 95% CI 2.23, 4.77). This study is the first one reporting direct economic burden on Chinese ALS patients. Efforts should be made to develop cost-effective diagnostic tools in order that sources of medical cost were more effectively allocated, and this disease was detected earlier.


Introduction
Amyotrophic lateral sclerosis (ALS), also known as a motor neuron disease or Lou Gehrig disease, is a specific neurodegenerative disease that progressively causes the death of upper and lower motor neurons controlling voluntary muscles. To our knowledge, ALS is an incurable disease, characterized by stiff muscles, muscle twitching, and gradual worsening of weakness. ALS had a poor prognosis and high mortality with patients died Editor: Narayan Subramanian.
HS and J-CL contributed equally to this work. The data that support the findings of this study are available from Directorate of Medical Services, Logistics Support Department of PLA, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of Directorate of Medical Services, Logistics Support Department of PLA.
The data that support the findings of this study are available from a third party, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are available from the authors upon reasonable request and with permission of the third party. within 3 to 5 years due to respiratory paralysis. [1] The incidence of ALS was estimated to be 1.9/100,000/yr worldwide, [2] whereas in China, the annual incidence or prevalence of ALS is lower than developed countries, such as Japan, Europe, or Australia. [3] It is estimated that number of ALS patients will increase 69% from 2015 to 2040 across the world. [4] ALS is relatively rare, yet imposing considerable economic and social burden. A recent study reported that its annual total cost per patient ranged from US$ 13,667 in Denmark to as high as US$ 69,475 in the United States. [5] Another study in the USA found monthly costs per patient increased 9 months before diagnosis, peaked at the index month (Medicare: US $10,398; commercial: US $9354) and decreased but remained high post-index. [6] Diagnosis of ALS is primarily based on the symptoms and signs after ruling out other diseases. In the United States, $10,000 to $20,000 per patient was needed for a neurologist to get a confirmative ALS diagnosis. [7] However, economic burden on Chinese ALS patients has not been reported. Phenotypic differences, like mean age-at-onset or diagnostic delay, were found among German and Chinese ALS patients. [8] In addition, clinical characteristics and outcomes of Chinese patients with sporadic ALS were quite different compared with patients from other countries. [9] These findings may indicate the sources of differences in economic burden between Chinese ALS patients and other groups.
Further, cardiovascular, metabolic, or neuropsychiatric comorbidities of ALS, make it difficult to isolate the specific contribution of ALS to medical economic burden. [5] Thus, we aimed to investigate medical cost and healthcare utilization for Chinese patients with ALS in the study where the control group had similar comorbidities, and under the correspondingly the same medical insurance system [Urban Employees (UE), Urban and Rural Residents (URR), Free Medical Care (FMC), and Self-Pay Medical Care (SPMC)]. The data set, derived from all military hospitals in China, provided an edge in studying rare disease, like ALS, as it includes over 20 million people, representing a possibly largest scale of Chinese ALS patients of all ages. The estimated medical cost and healthcare utilization will provide insight into offering reasonable amount of financial support for Chinese ALS patients and will improve allocation and efficient use of medical resources.

Data resource
A retrospective analysis was done with medical big data collected from all Chinese military hospitals, which serve not only the military personnel, but also civilians, taking up a significantly high proportion of Chinese health services. [10] Hospitals in this study are mostly located in key cities of provinces. The database includes patients' demographic information, vital signs, diagnoses of diseases, medical orders, examination reports, lab tests results, drugs, medical supplies, or even infections. The present study had analysis performed with historical de-identified data; thus, it was exempt from Institutional Review Boards (IRB) Committee of Fourth Military Medical University.

Study population
Study population comprises patients with a first major diagnosis for ALS by referring to the International Statistical Classification of Diseases, and Related Health Problems (ICD-10) G12.2 (motor neuron disease). We excluded subjects diagnosed with other systemic atrophies including G10 (Huntington disease), G11 (Hereditary ataxia), G13 (Systemic atrophies primarily affecting central nervous system in diseases classified elsewhere). The Day 1 for hospitalization of ALS patients was defined as the index date.
A total number of 880 inpatients of all military hospitals from 2015 to 2018 were included as subjects by nonprobability sampling using structured query language expression of Oracle databases. Records lacking in any primary keys, including hospital information, patient's identity, and visit times, were excluded. The control groups were set with criteria that patients should have at least one hospitalization visit and no diagnosis code for ALS or other nervous system diseases. The controls were match on age, gender, ethnic group, geographic region, length of stay, year of diagnosis and comorbidity. Figure 1 illustrates the detailed scheme.

Covariates and outcomes of interest
The ALS subjects and controls were matched on these comorbidities to maximize the association of medical cost and health care utilization with ALS. The structures of medical cost, outpatient visits, hospitalizations, emergency department visits, laboratory tests, imaging [Electromyography (EMG), nerve conduction velocity (NCV), Ultrasound, Computed Tomography, and Magnetic Resonance Imaging (MRI)] were selected as outcomes of interest. We measured the outcomes at 2 time points in reference to the index date: 1 year before, 1 year after.

Statistical analysis
Univariate descriptive analysis was employed to described quantitative and qualitative variables as count or median. A collection of summarization techniques has been formulated as bubble chart, scatter plot by groups, scatter plot on maps. We compared outcomes of interest between ALS patients and controls on 1 year before, 1 year after their diagnosis date. Medical cost structure was analyzed based on all details of medical expenses. Healthcare utilization was identified as events and Odd Ratio (OR) (events 1 year after ALS diagnosis/controls/ events 1 year before ALS diagnosis/controls). Without using samples to infer the population, we showed current status of Chinese ALS patients through statistical description. Data were analyzed with SPSS version 24.0, illustrated using GraphPad Prism 6, and Python (version 5.1.0; Anaconda3).

Results
We identified 3946 people with a first diagnosis of Spinal Muscular Atrophy and Related Syndromes (G12) from 2015 to 2018, of which 3447 were excluded for lack of medical records 1 year before and 1 years after the index diagnosis. The remained 480 people with ALS (G12.2) and 400 non-ALS patients, hitting a total of 880 people (Fig. 1) as the study population, were matched by age, gender, ethnic group, geographic region, length of stay, year of diagnosis.

Demographic characteristics
The mean ± SD (range) age of ALS patients in our study was 57.37 ± 15.09 (7-92) years old at the index date. Male or female  (Table 1). Interestingly, in terms of ABO blood type, type-O took the highest proportion, 20.21% in ALS groups, in contrast to a mere 7.00% in control groups. As rare blood types, proportions of Rh-negative in ALS patients (10.63%) were far higher than that of the controlled cases (0.50%).

Medical costs in ALS
Within the specific three-character category G12, we compared the number of patients and corresponding medical cost between its subcategories, where G12.2 (motor neuron disease) took the largest proportion of the medical expenses, 12,852 CNY ( Fig. 2A). Further, we carried out an analysis on geographical characteristics of ALS patients via their birthplace codes in county areas in China, which showed that major ALS patients came from Shaanxi (65), Chongqing (44), Hebei (37), Shandong (28), Jiangsu (23), Sichuan (21) (Fig. 2B). As for medical cost per patient, Guizhou, Guangdong, Henan reported higher medical cost than the rest provinces and regions, with a total of 117,665 CNY, 85,040 CNY, and 69,281 CNY, respectively (Fig. 2C). Figure 2D shows the annual total medical costs per patient among ALS patients and controlled patients during peridiagnosis period (1 year before vs 1 year post  (Fig. 2E). For reference, the annual healthcare cost of non-ALS controls was almost constant between 1 year before (17,676 CNY (IQR 9518 CNY -42,856 CNY)) and 1 year post 17,498 CNY (IQR 8909 CNY -38,495 CNY) (Fig. 2D). We further analyzed the structure of medical

Healthcare utilization in ALS
On the basis of estimating medical cost, we further determined healthcare utilization parameters at the index date and 1 year before, respectively. Compared with the year before the index date, utilization parameters, like hospitalizations, EMG, NCV, MRI, increased markedly during the following year among ALS patients, which were not observed in controlled group.

Discussion
Investigation on the data retrieved from a large military medical database show that medical costs for ALS patients were close to that of the non-ALS control group under similar conditions. However, the total annual cost increased about 2-fold during the ALS diagnosis year, far higher than the controls, among which medical costs during the ALS diagnosis year reached 10,170 CNY, 15,734 CNY, 41,253 CNY, 26,083 CNY with URR, UE, FMC, or SPMC, respectively. As major sources of medical costs, drug usage took a high proportion of economic burden for ALS patients. As the healthcare utilization encompassed hospitalizations, laboratory tests, EMG, or NCV test, of its increment may, to some extent, explain the increase of medical costs.
The onset of ALS is insidious and to our knowledge, it is difficult to get definite diagnosis with single diagnostic tool, therefore physicians always determined the ALS based on symptoms and signs. [11] Differential diagnosis were needed to rule out diseases, such as multifocal motor neuropathy or multiple sclerosis, whose symptoms closely resemble ALS. [12] Compared to the 11 months (range: 6-21) diagnostic delay from ALS onset in a cohort of 260 patients in Italy, [13] median diagnostic delay in China was 14 months, [9] indicating a relatively high medical costs during outpatient visits or hospitalizations in the extended process for ALS diagnosis indicated.
The mean age or male to female ratio was consistent with previous results. The mean age at onset was 52.4 years at 10 ALS centers of Chinese ALS Association from March 1, 2009 to August 31, 2009. [14] The male to female ratio was 1.69:1 among 2101 ALS patients collected in the University Third Hospital, Beijing, China. [15] To our knowledge, practical diagnostic biomarkers in the blood or cerebrospinal fluid has not been developed for the ALS disease. [16] As rare blood type, the proportion of Rh-negative patients in ALS groups were about 20fold compared with the matched controls. In previous report, no association of specific blood groups was found between ALS and controls. [17] The differences may be derived from the study population, or the study design. Though there is no proper account for this high constituent ratio of type-O or Rh-negative type yet, these findings may contribute potentially to the early detection of ALS in Chinese patients.
As a relatively rare disease, the most cost-effective modality to predict the severity of ALS has not been clearly illustrated. In this study, ALS treatment was examined by referring to existing medical practices, like EMG, NCV, or MRI test. To be specific, EMG, as key differential diagnosis tools, was used to detect electrical activity in muscles, identifying loss of lower motor neurons; [18] EMG abnormalities were observed in approximately 40% of asymptomatic limb muscles in 150 Chinese ALS patients. [19] NCV, another diagnostic tool to assess peripheral nerve conduction function, were used to exclude the possibility of peripheral neuropathy or myopathy. [20] The computed tomography and ultrasound tests increased slightly during ALS diagnosis compared with controls, implying that some examinations may have been used unnecessarily. Thus, the medical cost and corresponding utilizations during the first diagnosis of ALS suggested that more efficient tools of screening and assessment were needed.  This study also bridged some gaps in the existing literature. First, until our study, no direct economic burden for Chinese ALS patients from a large medical database was reported. Unlike in USA, [21] Ireland, [22] and Canada, [23] annual medical costs per Chinese ALS patient with 4 system, URR, UE, FMC, or SPMC were far lower than western developed countries, reflecting the gap of economic level. Second, by establishing matched controls with similar conditions as references, we eliminated confounding factors driven by other complications. Third, we retrieved medical record during outpatient visits or hospitalizations before the first ALS diagnosis and made it possible to analyze variances of cost and healthcare utilizations brought by this disease.
This study has some limitations. There may exist population selection bias, for patients with ALS who once visited civilian hospitals, are not captured in this medical database. Besides, economic burden outside hospitals, such as assistive devices or caregiver burden, has not been calculated into the overall medical burden of ALS. Third, we were not able to distinguish among

Conclusions
In summary, this study reported current status of economic burden for Chinese ALS patients. Currently, no effective cures for ALS or treatments were found to stop disease progression. From perspective of medical big data development, efforts should be made to develop cost-effective diagnostic tools so that this disease was detected earlier and economic burden was reduced for each ALS patient.