Estimating Disability Weight of Human Neurocysticercosis in Dali, Yunnan province, China

Background Human cysticercosis, caused by the liver parasite Taenia solium, remains an important neglected tropical disease in China. In Yunnan province, a large proportion of in-patients with cysticercosis suffer from neurocysticercosis. Neurocysticercosis, though preventable, is a cause of epilepsy globally. Disability weight is an important parameter when estimating neurocysticercosis disease burden and assessing disability-adjusted life years. However, there is a paucity of information on disability weight in patients with neurocysticercosis. Methods was compared using t-test or analysis of variance when data were normally distributed. Bonferroni test was used to make pairwise comparisons between groups when the difference was statistically signicant. Mann-Whitney U test or Kruskal-Wallis test was used to compare DW for non-normally distributed data. Bonferroni test was performed when the difference was statistically signicant.


Background
Human cysticercosis is a tissue infection caused by larvae of the parasite Taenia solium [1][2][3][4][5][6]. The Food and Agriculture Organization (FAO) categorized cysticercosis as a neglected zoonotic disease in 2014 after it was classi ed by the World Health Organization (WHO) as a neglected tropical disease in 2010 [7][8][9]. Cysticercosis has been identi ed as a major health problem in Asia and is also endemic in developing countries of Latin America and Africa [1]. In addition, it has also been recently identi ed as an emerging infectious disease in some developed countries due to increasing international communications [2,10]. The World Health Assembly passed the WHA66.12 resolution in 2013, targeting elimination of cysticercosis as a public health problem [11]. Cysticercosis can damage any tissue in the human body, including the central nervous system, subcutaneous tissues, eyes, and muscles [1,2]. Symptoms vary by location. When the larvae invade central nervous system, it can cause neurocysticercosis (NCC). NCC accounts for the major proportion of human cysticercosis patients.
Clinical symptoms of NCC can result in serious health complications [4,5,7,12,13]. The parasite causes a large number of NCC-associated epilepsy and deaths, with 2.79 million disability-adjusted life years (DALYs) [14]. Following increasing global concern, disease burden studies have been carried out in many countries [9,[15][16][17]. In these studies, NCC disability weight (DW) is a key parameter in estimation of NCC burden [18]. However, the DW of epilepsy or headache was used to estimate NCC DALYs rather than that of NCC directly [9,16,17]. Moreover, not all NCC patients manifest symptoms of epilepsy or headache [4].
It is, therefore, imperative to obtain NCC DW in the research of disease burden.
Between 1987 and 1991, an inter-disciplinary, ve-country group developed the EuroQol instrument to measure and evaluate health status [19][20][21]. The ve-dimensional, three-level generic measuring instrument was subsequently termed the 'EQ-5D'. A cognitive dimension was later added to the initial questionnaire, making the tool EQ-5D plus or EQ-5D + C [19,22].
This study aimed to obtain NCC DW using EQ-5D + C tool in Dali, Yunnan province. Estimating NCC DW will not only provide basic scienti c evidence in cysticercosis disease burden research, but also inform control strategies in this area.

Study area
This study was carried out in the Dali Prefectural Institute of Research and Control on Schistosomiasis, located in the Dali Bai Autonomous Prefecture, Yunnan, China. Dali prefecture is located in the central western part of Yunnan, 350 km northwest of Kunming, the capital of Yunnan province. It has a territory of 29,459 square kilometres and a population of 3,400,000. The prefecture is subdivided into 12 countylevel divisions: one county-level city, eight counties, and three autonomous counties. In China, Dali is a hotspot for cysticercosis endemicity due to the local dietary habit of eating undercooked pork. For years, the Dali Prefectural Institute of Research and Control on Schistosomiasis has served as a reference hospital for cysticercosis patients in Yunnan province.

Study design
Participants were separated into 2 groups, the rst-visit group and the follow-up group. The rst-visit group were those patients who received standard anti-cysticercosis treatment for the rst time in the hospital while the follow-up group were those who received subsequent treatment in the hospital. A survey was administered to both groups in 2018. It consisted of three parts: part one, demographical characteristics including age, sex, occupation, ethnic status, education, and marriage status; part two, case diagnosis and treatment information, including diagnosis basis, case classi cation, previous medical story, and clinical manifestations; and part three, the EQ-5D + C survey, which included 6-domains and a visual analogue scale.

Selection of participants
Participants were inpatients in the Dali Prefectural Institute of Research and Control on Schistosomiasis from September 2017 to September 2018. According to the national standardized diagnostic criteria for NCC (WS381-2012) [23], patients were classi ed into three categories, suspected cases, clinically diagnosed cases, and con rmed cases. Only con rmed cases were eligible for inclusion in the survey. Exclusion criteria were: (i) patients who did not nish the questionnaire, (ii) cysticercosis other than NCC, (iii) patients with other underlying diseases, (iv) mixed cysticercosis, (v) patients younger than 5 years old.

Sampling
The sample size (n) was determined using the statistical formula: Where α is the type I error rate (0.05) and α = 0.05 (two-sided), π is the population rate and if π = 0.5, the maximum sample size can be obtained. σ is the allowable error (0.1), N reaches to 96. Considering the cluster sampling power as 2, then the required sample size was 192. The study participants were then selected by systematic random sampling using selected participants as a sampling frame. As a result, a total of 210 participants were enrolled in the study (Fig. 1).

Data collection
Surveys were conducted face-to-face and data collected by trained investigators. Data were doubleentered and logic correction was conducted using EpiData3.1 (The EpiData Association, Odense, Denmark).

Statistical analyses
Descriptive analysis was used to show the demographic characteristics and clinical symptoms of NCC patients; the χ 2 test was used to compare the differences in clinical symptoms between the rst-visit and follow-up groups. Descriptive analyses were used to show the distribution of patients' six-dimensional and EQ-VAS scores. T-test or analysis of variance was used to compare the EQ-VAS scores for normally distributed data, while Mann-Whitney U or Kruskal-Wallis test was used for non-normally distributed data. The χ 2 test or Fisher's exact probability method was used to compare the proportion of health di culties in the six dimensions. DW was determined by the formula DW = 1-VAS/100, with VAS as the patient's EQ-VAS self-assessment. SPSS 20.0 software (IBM, Armonk, NY, US) was used for statistical analysis. DW was compared using t-test or analysis of variance when data were normally distributed. Bonferroni test was used to make pairwise comparisons between groups when the difference was statistically signi cant. Mann-Whitney U test or Kruskal-Wallis test was used to compare DW for non-normally distributed data. Bonferroni test was performed when the difference was statistically signi cant.

Socio-demographic ndings
A total of 210 participants with age range of 7 to 74 years old were included. The mean age was 38.9 ± 15.5. The largest number of study participants, 69 (32.9%) was from the age group 30-44 years. The sex ratio (M/F) was 1.9 with 138 males and 72 females. Bai was the predominant ethnic group (29.1%), followed by Yi with 17.2%. Out of the 210 participants, 84 were in the rst-visit group and 126 were in the follow-up group. Respondents' demographic and socioeconomic characteristics are summarised in Table 1. Clinical symptoms The most prominent clinical symptoms in the rst-visit group and follow-up group were headache (64 cases) and hypomnesis (85 cases), respectively. Headache, epilepsy, and hypomnesis were the most frequently reported manifestations by the rst-visit group while hypomnesis, headache, and muscle paraesthesia were more common in the follow-up group. The percentage of epilepsy was 54.76% in the rst-visit group compared to 25.40% in the follow-up group (χ 2 = 18.52, P < 0.001), and headache was 76.19% in the rst-visit group compared to 62.70% in the follow-up group (χ 2 = 8.52, P < 0.05), respectively. However, the proportion of patients with hypomnesis was signi cantly higher in the followup group (67.46%) than that in the rst-visit group (52.38%; χ 2 = 8.76, P < 0.05)( Table 2). In terms of the six-dimensional distribution, pain and/or discomfort ranked highest (77.38%) in the rstvisit group while cognition was highest in the follow-up group (74.60%). Compared to those in the rstvisit group, health barriers were less frequently reported in all six dimensions in the follow-up group, except for cognitive dimension (Table 4). Note: * Fisher exact probability method; a "Problem" = moderate or extreme health barriers The ndings indicated that age, occupation, usual activities, pain/discomfort, anxiety/depression, and cognition were predisposing factors for the EQ-VAS score (Table 5). Multiple linear regression analysis was applied by using EQ-VAS score as the dependent variable. The results of a multiple linear regression model showed a signi cant difference (F = 16.99, P < 0.001). R 2 was equal to 0.46, indicating that the variables included could account for 46% of the total DW variation. Multivariate analysis found that EQ-VAS score was negatively correlated with daily activities, pain/discomfort, anxiety/depression, and cognitive ability. The standardized regression coe cients showed that cognitive ability had the greatest impact on EQ-VAS score (-0.32), followed by daily activities (-0.27), and pain/discomfort (-0.25) ( Table 6).

Disability weights
According to the normality test, DW did not conform with the normal distribution (Shapiro-Wilk test W = 0.73, P < 0.001). The DW in the rst-visit and follow-up groups were both 0.3 (0.2-0.4) ( Table 7, Table 8), without statistical signi cance (Z = 0.74, P > 0.05) ( Table 9). No signi cant signi cance was found in the proportion of gender (χ 2 = 2.87, P > 0.05) or age (χ 2 = 2.08, P > 0.05) between the two groups. Therefore, DWs in both groups were combined and analysed as a whole. The overall DW of NCC was 0.3 (0.2-0.4), without signi cant difference between genders (Z=-0.15, P > 0.05). However, DWs were not equal among age groups (χ 2 = 13.25, P < 0.05). DW tended to increase with age. Pairwise comparison showed no signi cant differences between the age groups except for the 5-14 age group (Table 10).

Discussion
Cysticercosis has historically been widely distributed in China [3,10,[24][25]. However, the prevalence has recently decreased gradually in parts of southwestern China due to improvements in social and economic development. Nevertheless, a large proportion of residents, especially in Yunnan province, are accustomed to consuming raw or undercooked pork [3,13,[25][26][27].
Previous studies showed that seizures were the most common clinical manifestation among NCC patients [4,6,13,[28][29][30][31][32]. However, the main clinical symptom among patients in this study was headache, followed by hypomnesis and seizure. This may be because the period of epilepsy symptoms in this study was de ned as whether there was epilepsy in the last month, rather than whether there was epilepsy at any time prior. In addition, some of the patients received standardized treatment, indicating that treatment may effectively alleviate symptoms of seizures [5,[33][34][35].
The EQ-5D + C is a valid tool widely used to measure health-related quality of life [20,[36][37][38]. It was found that pain/discomfort and cognitive disorders were the most prominent symptoms attributable to Many studies have reported NCC DALYs globally, but few provide information on NCC DW [9,[14][15][16][17]46].
This study showed a way to report NCC DW using EQ-5D + C. The general NCC DW was 0.30 in both the rst-visit and follow-up groups. Except for hypomnesis, other clinical symptoms were alleviated in followup patients. Nevertheless, there was no statistical signi cance of DW between the two groups. This may because that treatment did not help improve the patients' subjective health-related quality of life. Result showed that cognitive ability has the greatest impact on EQ-VAS score. In general, cognitive ability include memory, observation, and imagination [47]. Compared with rst-visit patients, hypomnesis may be responsible for its DW in follow-up patients. The WHO [48] de nes health as a "state of complete physical, mental, and social well-being, and not merely the absence of disease or in rmity". With this in mind, changes in clinical symptoms alone cannot effectively improve a patient's overall health. It is recommended that psychological treatment be incorporated into the treatment strategy for NCC patients.
Therefore, it is proposed that in treating NCC patients, psychological counseling should be considered to help reduce emotional burden and psychological pressure caused by memory loss.
Based on the EQ-VAS score, patients rated themselves after getting to know their own health status. However, this method is highly subjective and may easily affected by the patients' mood and other factors, which may lead to unstable result [49].

Conclusions
The health-related quality of life of NCC patients was seriously impaired, and cognitive competence is the most prominent health barrier. Psychological medical care should be taken as one component in the treatment of NCC patients.