Analysis of the Situation of Unmet Health Care Need Among Extremely Rare Disease of Gaucher

The diagnosis and health care of patients with rare diseases present a tremendous challenge worldwide. This study estimated the health service utilization, cost of illness, and patients with Gaucher disease (GD)’s/caregivers’ health-related quality of life in China. An online retrospective survey of patients with GD and their caregivers was conducted during May-June 2018. Socio-demographic, health service utilization, disease-related expenses, social support, sleep quality (Pittsburgh Sleep Quality Index [PSQI]), and the Short Form Health Survey (SF-36) were investigated. Using self-reported information, we estimated the annual cost of illness, including direct medical, direct non-medical, and indirect medical costs.

The pro le of health service utilization of patients with GD are shown in Table 2. Patients' original reasons for visiting the doctors included enlarged liver and spleen (71.4%), normocytic anemia (36.7%), dyskinesia (12.2%), and osteopenia (8.2%). Among all surveyed patients, the two most frequent symptoms were hepatosplenomegaly (85.7%) and normocytic anemia (73.5%). Many patients (43.5%) reported that they had to travel to the tertiary hospitals in other provinces to get con rmed diagnosed.

Cost of Illness
The cost of illness of patients with GD is shown in Table 4 and Fig. 1  The average annual direct non-medical costs added up to $4,974.4 (10.0% of total costs). With both direct medical and non-medical costs, the annual direct cost was estimated at $46,790.6. The indirect costs for GD primarily originated from the productivity loss to the caregivers while caring for patients with GD (4.0% of total costs). Accordingly, the total annual indirect costs were estimated at $3,134.2.

Social Support and Health-Related Quality of Life of Patients and Their Caregivers
The total Social Support Rating Scale (SSRS), Pittsburgh Sleep Quality Index (PSQI), and Short Form Health Survey (SF-36) scores for patients with GD, their caregivers, and norms of healthy Chinese people are shown in Table 5. Patients with GD and their caregivers reported mean SSRS scores of 25.3 ± 5.7 and 26.1 ± 5.6, respectively. Both were remarkably lower than the SSRS score (40.5 ± 6.8) for healthy Chinese people (all p-value < 0.05). The mean PSQI scores for patients and caregivers scores were 7.9 ± 2.9 and 8.7 ± 3.6, respectively. Patients with GD and their caregivers reported two times higher PSQI scores than the normal values for healthy Chinese people (3.9 ± 2.5) (all p-value < 0.05). Overall, 16 (32.7%) of caregivers rated sleep problems using PSQI dimensions. All the seven PSQI dimension scores for the caregivers were higher than the normal values for healthy Chinese people (all p-value < 0.05).

Discussion
There is a dearth of evidence on the health service utilization and economic burden of rare disease patients with GD in China. In this study, we performed a comprehensive, analysis and identi ed a wide range of unmet needs and problems of patients with GD in their health service utilization, cost of illness, and health-related quality of life associated with GD in China. Families of patients with GD often experienced delays in diagnosis and misdiagnosis, enormous economic cost, and caregiving burden. Lack of social support often resulted in deteriorated health related quality of life and other major health issues which deserve great attention of society.
Misdiagnosis or delayed diagnosis is especially detrimental to patients [20]. Although several publications have acknowledged that diagnostic delays or misdiagnosis often occurs in obtaining a de nitive GD diagnosis [21], this study further quanti ed the exact time delay and di culty in obtaining the correct diagnosis that occurred in 79.6% surveyed patients. 59.2% of them experienced misdiagnosis, of up to 5.0 ± 9.6 times. Besides, patients reported multiple referrals (3.9 ± 3.1) to different hospitals before their disease was accurately determined. Studies indicated the process of continually seeking diagnosis was a traumatic experience for patients with rare diseases [22]. Patients spent 1.  [21,23,24], and preventing psychological stress for the family [25].
The main reason for the prolonged diagnostic delay most likely is insu cient knowledge on GD [26][27][28] by the medical community due to its low prevalence, extreme variations in clinical manifestations [23,29], and severity of symptoms (e.g., bone pain/bone crisis, thrombocytopenia, and splenomegaly) [12,21,23]. Nevertheless, a survey revealed that only one of ve hematologist-oncologist considers GD in the differential diagnosis of patients with a history of anemia, thrombocytopenia, hepatomegaly, splenomegaly, and bone pain. In this study, 79.6% of patients had visited doctors with obvious symptoms (osteopenia, normocytic anemia, dyskinesia, and obviously enlarged liver) in whom GD was not diagnosed in a timely manner. Therefore, establishing an information center of rare disease symptoms [30] and providing training on diagnostic methods and appropriate clinical guidelines to medical doctors are essential to effectively support early diagnosis of rare diseases. Besides, because most rare diseases are caused by DNA mutations or are recessive genetic diseases [31], family history and genetic screening could facilitate earlier diagnosis, while genetic testing can be used to con rm disease diagnosis [32]. Nine families reported a family history of the same disorder in this study, indicating a need for awareness of genetic disorders to highlight the importance and burden of genetic diseases. Therefore, newborn screening for disorders that meet the criteria for population screening [33], use of reliable diagnostic tools [1,[34][35][36], raising of awareness on the early signs/symptoms [37] or less severe manifestation [21], and initiation of therapy early for rare diseases [30], are essential.
Besides, other studies also shown a signi cant proportion of patients (e.g., asymptomatic or mildly affected) never accessed medical attention or required treatment [41,42]. In this study, even 20 (40.8%) of patients did not clear the type of their GD. Maaswinkel Mooij et al. [43] found that no treatment for type 1 GD with enzyme replacement therapy can worsen the condition at any age in non-Jewish patients [44]. Besides, 4 of 9 patients who took imiglucerase interrupted their treatment in this study. However, Drelichman et al. emphasized the need to avoid interruption of medical treatment of patients with GD because of recurrent organomegaly, growth delays, and skeletal manifestations [45]. Furthermore, a previous study identi ed association of high immunoglobulin (Ig) A and IgG levels with long-term complications [46]. Therefore, long-term treatment and surveillance are required for improving the e ciency of GD management [47].
To estimate the annual cost of GD, we presented the core components of costs for health care services separately, from the payer' perspectives, excluding patients' out-of-pocket costs and informal care.  [19].
Therefore, further efforts are needed to address the issue of pharmaceutical treatment costs holistically [51].
However, both the availability and affordability of orphan drugs [52] for the treatment of rare diseases in China are low [17]. Poor accessibility to drugs is the most problematic issue for patients with rare disease in China [53]. In this study, 89.8% of patients perceived the availability of therapeutic drugs to be problematic. Furthermore, 35.0% of patients who received pharmaceutical treatment could not take the medication as prescribed. These gures demonstrate poor availability of rare drugs. Several hindering factors contributing to the low availability of orphan drugs in China include the low market availability of orphan drugs in China [17,54]. Compared with the earliest launch time globally, the average delay in the market authorization of orphan drugs for rare diseases in China was 7.7-9 years [17,54]. Imiglucerase received China's marketing authorization in 2008 [17]. On May 22, 2018, the Chinese Government o cially included GD on the rst list of rare diseases [55]. With a low availability of < 30% [17], many Chinese patients currently pay out-of-pocket for international treatments that are currently unapproved in China [56]. Second, lack of research and development and supply incentive policies for orphan drugs are possible reasons leading to low market availability [56]. Under a market-oriented economy, due to a small market share of orphan drugs, most pharmaceutical manufacturers are unwilling to invest in research and development on the production of orphan drugs without an incentive policy [17]. Third, low public hospital availability of orphan drugs is another major issue both for the patients [57] and the Chinese government. In this study, 31 (77.5%) of patients reported purchasing medicine at pharmacies or overseas. Due to the unique market attributes (e.g., low market volume, low pro t, low turnover rate, and high price) of orphan drugs, and as the reform of public hospitals in China stipulated (control the share of drug sales in total hospital revenue), hospitals have little incentive to stock and prescribe expensive orphan drugs [58]. Lastly, there are no public national or provincial networks for rare diseases or orphan drugs to share useful information about these diseases, e.g., treatment or supply information. Therefore, it is urgent to improve the availability of rare drugs, e.g., simplify the approval procedure for imported orphan drugs, increase research and development investment, formulate incentive policies, and establish information-sharing platforms.
Furthermore, to recoup research and development costs and for pro t, orphan drug manufacturers often adopt a strategy of charging high prices [59]. Miglustat and imiglucerase, which are effective medications for treating GD, had annual costs of $116,800 and $140,200, respectively in 2004 [60], and the expenditure share was > 3% of per capita GDP [16]. A study focusing on seven selected rare diseases found that the affordability of treatment was relatively poor, with the health expenditure for GD equivalent to 69.34 years per urban resident's income in 2014 [16]. In this study, 49 (100%) of patients reported the unaffordability of therapeutic drugs. A few cities/provinces are active in creating rare disease lists with a high reimbursement at 80%-95% [61]. However, due to the restrictive reimbursement caps, even with 5% out-ofpocket expenses, few drugs could be afforded by high-income urban residents [62]. The out-of-pocket costs are still unaffordable for many patients. Therefore, the low affordability of orphan drugs may be closely associated with many factors, e.g., high drug prices, lack of insurance coverage, low reimbursement rates, and low-income levels for Chinese residents [17]. Several measures should be taken to improve the affordability of orphan drugs in China including: 1) formulating and implementing incentive policies [59,63] to promote the development and supply of generic drugs, and to control the price of orphan drugs [64,65]; 2) developing an orphan drug reimbursement system to increase insurance coverage; and 3) developing government-supported programs for patients with rare disease.
Due to the delayed diagnosis, unavailability, and unaffordability of costly drugs, 79.6% of 49 patients with GD felt poorly con dent about future treatment. Moreover, the total social support, sleep quality, and quality of life scores for patients with GD/caregivers were signi cantly lower than those for normal healthy Chinese people. Thus, patients with GD experience limited access to health services, substantial costs, and low health-related quality of life [66].

Limitations
First, this study is limited primarily by the small sample size. Due to the rarity of GD, data collection was challenging. Second, families who voluntarily participate in the GD patient registry may represent a more compliant and motivated patient cohort in general. Thus, the ndings may not be generalized to all.
Thirdly, this study relied on responders' recall of health service utilization and the unit costs to capture a more inclusive set of cost components usually not included in billing data, such as tra c costs, formal care costs, and productivity loss to caregivers. To enhance the accuracy of the estimates reported, patients with GD and caregivers were asked to complete questions regarding the average consumption amount and costs per resource in the past year. However, recall bias may have led to errors. Lastly, utilization and corresponding costs from pharmaceutical use of prescription were not included in the survey. For pharmaceutical costs, we asked for the "annual purchase of imported/ domestic medicines," which could have resulted in an underestimation of this study's GD's nancial burden. Hence, appraisal costs using data from health insurance is needed.

Conclusions
Patients with GD often encountered the frustrating experience of high misdiagnosis rate, long-delayed diagnosis, substantial costs, and deteriorated health-related quality of life in China. For patients with GD, a rapid, precise diagnostic tool for earlier and timely de nitive diagnosis of GD is urgently needed. A more tailored medical insurance program that can effectively address the unmet health need of patients with GD should be designed. Measures targeting improving the availability and affordability of orphan drug needs to be developed holistically in China. Furthermore, newborn screening for disorders that meet the criteria for population screening with reliable diagnostic tools should be explored. Besides, enhancing doctors' ability in GD diagnosis and differential diagnosis, raising awareness of the early signs/symptoms or less severe manifestation to achieve earlier diagnosis and timely therapy of rare diseases, relieve caregivers' burden, and provide much needed social support are essential. This research can facilitate greater societal awareness of rare diseases, help policymakers develop appropriate intervention programs, and inform healthcare-focused support schemes and policies for patients and their families.

Development of the Survey
An online retrospective survey of patients with GD and their caregivers was conducted for data collection.
The inclusion criteria for patients were as follows: (i) a diagnosis of GD by a physician before the study; (ii) with a caregiver who was familiar with the whole treatment process of them; (iii) if the patient was under 15, with a voluntary caregiver to help answer all the real information of them. The inclusion criteria for the caregivers included that he/she was: (i) the primary caregiver; (ii) was thoroughly familiar with the patient's disease; (iii) accompany patients every visit to the treatment; and (vi) was able to understand the content of the questionnaire.

Socio-demographic characteristics
The patients' socio-demographic characteristics included age, sex, current residence, mean age at onset, disease duration, type of GD, family history of GD, and caregivers' number. Socio-demographic characteristics of the caregivers included the daily care time, experience of being a caregiver, stopped working because of patients' GD, change of weekly working scheme because of patient's GD, and members with chronic disease > 60/<5 years of age. Besides, whether living on minimum subsistence allowance or relatives' relief or not was also reported.

Medical service utilization and caregivers' perception of GD treatment
The medical service utilization data were collected in this study to assess patients' diagnosis and Additionally, caregivers' perceptions et al., "the availability and affordability of therapeutic drugs," "di culties during GD treatment," "knowledge of GD," and "di culties to get diagnosis information" were also investigated.

Cost of illness
We retrospectively evaluated the total cost of GD, comprising direct medical, direct non-medical, and indirect costs for individual patients and their families. Costs comprised the following components: Direct Respondents were also asked to recall the use of resources: (i) outpatient/inpatient visit frequencies (consultations) annually; (ii) the average days for each outpatient/inpatient visit; and (iii) total days of lost wages annually. The resources used were multiplied by the corresponding unit cost to estimate the annual cost per patient, and the monetary values of the unit prices were reported in USD ($).

Zarit Burden Inventory
The subjective burden of the caregivers was measured by ZBI (22-item version), with item scores ranging from 0 (never) to 4 (nearly always). The total score ranges from 0 to 88, and scores < 21 correspond to little or no burden while scores > 61 correspond to severe burden [67].
Social Support Rating Scale SSRS comprise 10 items and includes three subscales: subjective support (4 items), objective support (3 items), and utilization of support (3 items). The total SSRS score ranges from 12 to 66 points, and higher scores indicate higher level of social support [68].

Pittsburgh Sleep Quality Index
Sleep quality of individuals in 1 month was measured using the PSQI (comprises 19 items and 7 dimensions), which includes subjective sleep quality, sleep onset latency, total sleep duration, sleep e ciency, sleep disturbances, use of sleep medication, and daytime dysfunction. The sub-total of the scores of each dimension ranges from 0 to 3, and the maximum score is 21. The cut-off score for PSQIde ned cases of poor sleep quality is ≥ 6. A Chinese PSQI version has been validated with adequate reliability [69].

The medical outcomes study 36-item Short Form
Quality of life was measured by the SF-36 comprising 36 items including the physical and mental component scores [70]. Subscale scores are then transformed from the normal scale to a 0-100 standardized score scale, with higher scores indicating more positive health status and a better healthrelated quality of life.

Statistical Analyses
Descriptive analysis was conducted to evaluate the social demographic characteristics, the status of medical service utilization, cost of illness, and health-related quality of life. Continuous variables were presented as means ± SDs, and categorical variables were presented as absolute and relative frequencies.
The 2.5 and 97.5% quantiles of the bootstrap distribution were used as the limits of the 95% CI. All statisitical analyses were conducted using SAS software, version 9.4 (SAS Institute, North Carolina, US