Bulletin of the World Health Organization

Assessment of the age-specific disability weight of chronic schistosomiasis japonica

Tie-Wu Jiaa, Xiao-Nong Zhoua, Xian-Hong Wanga, Jürg Utzingerb, Peter Steinmannb, Xiao-Hua Wua

Introduction

Schistosomiasis japonica is caused by the trematode Schistosoma japonicum, which is the only human blood fluke that occurs in China. Although significant progress has been made in controlling schistosomiasis in China, the disease remains of considerable public health significance, particularly in lake and marshland regions.13 The other two main human schistosome species are S. haematobium and S. mansoni.4 An infection with schistosomes is often followed by acute Katayama fever or nonspecific symptoms. If left untreated, an infection develops into a chronic condition characterized by hepatosplenic disease and impaired physical and cognitive development.1 The Global Burden of Disease study estimated the burden of schistosomiasis at 1.7 million disability-adjusted life years lost in 2002, mainly as a result of morbidity.5,6 However, recent reports and a meta-analysis suggest that the global burden of schistosomiasis might be several times higher.79 This discrepancy might be explained by the paucity of community-based data on the distribution and severity of nonfatal health outcomes caused by schistosomiasis and the challenges of estimating schistosome disability weights that are species-specific.

The following observations prompted us to reassess the disability weight of chronic schistosomiasis japonica. First, the age-specific disability weights used in the Global Burden of Disease study were very low, namely 0.005 among those aged < 15 years and 0.006 among those aged ⩾ 15 years. Similar estimates have been attached to, for example, facial vitiligo.5,9 Second, the same disability weights were assigned for chronic schistosomiasis regardless of the causative agent. However, an infection with S. japonicum causes distinct and more severe morbidity than that associated with other species, which is partially explained by the significantly higher egg production capacity of female S. japonicum worms.10 Third, while schistosome infection is the only sequelae of schistosomiasis considered in the Global Burden of Disease study, morbidity may persist long after parasitological cure. Finally, the Global Burden of Disease study did not consider other effects associated with schistosomiasis, such as anaemia, malnutrition, cognitive impairment and growth retardation.

The consequences of chronic schistosomiasis are many and various but are mainly governed by two aspects. First, there is direct morbidity resulting from pathological changes and clinical manifestations induced by the deposition of schistosome eggs in tissues and the subsequent inflammatory immune reactions and calcification of dead eggs. This may cause hepatomegaly, splenomegaly, hepatic fibrosis and various other symptoms, including abdominal pain, diarrhoea and blood in the stool. Second, additional morbidity also associated with schistosomiasis (for example, anaemia, growth retardation and cognitive impairment) is nonspecific and thus it may be difficult to tease apart from morbid sequelae caused by other diseases.9,11 Moreover, it may be difficult to identify chronic cases and to assess people’s health status in endemic settings where multiparasitism is the norm rather than the exception.1,12

The advent of portable ultrasound devices13 and progress made with questionnaires for self-reported health (for example, the EuroQol Group’s quality of life questionnaire known as EQ-5D)14,15 offer new avenues for obtaining community-based estimates of the disability weight caused by chronic schistosomiasis. Abdominal ultrasonography allows the direct visualization of pathology in the liver and spleen and facilitates the quantification of disease progression and resolution following treatment.13,1618 The EQ-5D questionnaire is a standardized instrument that measures people’s health status and their quality of life. In March 2000, a version of the EQ-5D questionnaire authorized by the EuroQol Group was validated in health-related quality of life studies in Beijing.19 The original questionnaire has been expanded to include a cognitive dimension; this version is known as EQ-5D plus.20,21 WHO has initiated a data collection strategy for health valuation that pays more attention to the individual notion of health in assessments of the burden of disease.22,23

The objective of our study was to estimate the age-specific disability weight of chronic schistosomiasis japonica by measuring health outcomes and overall quality of life among communities in schistosome-endemic settings in China.

Methods

Study area and population

This study was embedded in the third nationwide sampling survey on schistosomiasis, carried out in 2004.24 All tests and examinations adhered to standardized, quality-controlled procedures that are widely used in China. The sensitivity of the enzyme-linked immunosorbent assay (ELISA) used (Shenzhen Combined Biotech) was 97%.25 Technical and medical staff (for example, laboratory technicians and ultrasonographers) were well-acquainted with the procedures.

Our study was carried out between October 2004 and January 2005 in two schistosome-endemic counties. Dangtu county is located on the south bank of the Yangtze River in the eastern part of Anhui province. Hanshou county is economically less developed, and is situated in the marshlands of Hunan province’s western Dongting Lake region. A 3-km grid sampling technique was used to select villages for inclusion.26 Within each randomly selected grid cell, a single administrative village was chosen, based on its accessibility and representativeness in terms of schistosomiasis endemicity. Residents aged ⩾ 5 years were eligible for inclusion.

Questionnaire and diagnostic procedures

Participants who gave written informed consent were tested for schistosome-specific antibodies using an ELISA test.25 A positive test result indicates that a participant is infected with S. japonicum or has a recent history of infection (within the past 2 years).

All seropositive individuals were first interviewed using a standardized and pre-tested questionnaire to obtain information on sociodemographic variables (age, sex, educational attainment and occupation), treatment history, potential contacts with cercariae-infested water during the previous year and self-reported symptoms, using a recall period of 2 weeks. Next, functional outcomes were assessed using the EQ-5D plus questionnaire.14,15,20,21 This questionnaire obtains data on six dimensions: mobility, self-care, usual activities, pain or discomfort, anxiety or depression, and cognition. The attribute “usual activities” was clarified by mentioning specific areas of interest such as work, study and family or leisure activities. Attributes associated with “cognition” included memory, concentration, coherence and intelligence quotient (IQ). For each dimension, three possible outcomes were considered: no problems, moderate problems and extreme problems. The questionnaire also included a 20-cm visual analogue scale used for self-assessment of general health status. Participants are asked to rate their health on a scale from the best imaginable (100) to the worst imaginable (0).

Subsequently, an abdominal ultrasonographic examination was performed to assess the level and extent of hepatic morbidity and spleen-related morbidity. The ultrasonographic examination was performed on fasting participants (who had had no food 4 hours prior to the examination) positioned horizontally during relaxed inhalation. Pathology was graded according to standardized criteria.27,28 Hepatic fibrosis was graded from 0 (normal) to grade III (severe damage). Hepatomegaly was defined as a protrusion of the liver of > 3 cm under the xiphoid process or > 0 cm under the right costal margin at the midclavicular line. Splenomegaly was defined as a protrusion of the spleen of > 0 cm under the right subcostal margin at the midclavicular line.

Finally, the diagnosis of chronic schistosomiasis japonica was based on the following criteria: (i) positive serological result, (ii) self-reported contact with potentially schistosome-infested water, and (iii) either schistosomiasis-like hepatic fibrosis of grade II–III as detected by ultrasonography, or no or light hepatic fibrosis (grade 0 or I) with the presence of hepatomegaly or splenomegaly.27,29 Participants who met these inclusion criteria but had signs of portal hypertension due to hepatic fibrosis such as ascites, megalosplenia or dwarfism (advanced or late-stage schistosomiasis) were excluded. Although participants were not specifically examined for other diseases, those with clinical signs and symptoms clearly attributable to conditions other than schistosomiasis were excluded from further analysis.

Statistical analysis

All data were double-entered and validated using EpiData version 3.0 software. Statistical analyses were performed using SAS software version 8.1. The χ² test was used to examine differences between categories, and the Cochran–Mantel–Haenszel χ² test was used to explore linear associations between outcome variables and age. The disability weight for each individual was computed based on the self-rated health score on the linear visual analogue scale using the following formula: disability weight = 1 – (score/100); age-specific mean disability weights were also calculated. Analysis of variance (ANOVA) was used to test the differences in mean disability weights between age groups; the Bonferroni t-test was used for comparisons between groups.

In view of the hierarchical nature of the data, two multilevel regression models were developed with individuals assigned to level 1 and villages assigned to level 2.30,31 Model 1 explored the relationships between the questionnaire results and the disability weight, and model 2 assessed the correlation between schistosome-related morbidity and disability weight. In model 1, predictors included the sociodemographic data and the six dimensions of the questionnaire. In model 2, the independent variables included sociodemographic data, the number of previous antischistosomal treatments, reported symptoms, grade of hepatic fibrosis, and presence of hepatomegaly and splenomegaly. In both models, the county was considered to be a fixed effect. Independent variables showing no statistical significance (P > 0.05) were removed by a stepwise backward elimination procedure. Age, occupation, educational attainment and hepatic fibrosis were specified as categorical variables, with a reference category and a set of contrasted dummy variables. The two-level models had the following structure: yij = β0 + Σβkxkij + ui + eij, where yij is the dependent variable for subject j in village i, β0 is the intercept, xkij are the fixed variables (predictors), βk is the regression coefficient of the corresponding xk, ui is a random effect representing the variability between villages, eij is the residual or error resulting in assumed normal distribution with mean 0 and variance 1.30,31

Ethical approval and informed consent

The study protocol was approved by the institutional review board of the National Institute of Parasitic Diseases (Shanghai, China), and the WHO Research Ethics Review Committee. The objectives, procedures and potential risks were explained to county and village authorities, who subsequently informed the local population about the procedures. Written informed consent was sought from all participants. Seropositive individuals were treated with praziquantel, free of charge, according to Chinese national guidelines.27

Findings

Altogether, 91 villages were randomly selected; 40 in Dangtu county (including 15 villages that are non-endemic for S. japonicum) and 51 in Hanshou county (20 non-endemic villages). The estimated population of these villages was 134 385, of whom two-thirds had lived in their home village for at least 6 months in 2004. There were 77 387 people eligible for inclusion (aged ⩾ 5 years), and 59 765 were screened by ELISA; of these, 3405 (5.7%) tested positive. Complete questionnaire and ultrasound data were obtained from 2843 seropositive individuals.

Among those who were seropositive, the most common self-reported signs and symptoms were diarrhoea (30.0%; 853/2843), abdominal pain (23.0%; 655/2843) and blood in the stool (6.9%; 196/2843). Impaired capacity to work or study was mentioned by 18.7% (533/2843). Hepatic fibrosis was found in 57.9% (1646/2843; grade I: 36.7%, 1042/2843; grade II: 20.2%, 574/2843; grade III: 1.1%, 30/2843). The prevalence of hepatomegaly was 38.2% (1085/2843) and splenomegaly was 9.5% (269/2843).

In total, 1419 of the 2843 seropositive individuals (49.9%) met the criteria for chronic schistosomiasis japonica. Table 1 (available at: http://www.who.int/bulletin) shows the number and percentage of self-reported signs and symptoms as well as schistosome-related pathology, stratified by age, among all people with chronic schistosomiasis japonica. The highest proportion of chronic cases was observed among those aged 45 to 59 years, accounting for 43.2% (613/1419) of all chronic cases, whereas the proportion in the youngest age group – those aged 5 to 14 years – was 2.2% (31/1419). There were no statistically significant differences in the prevalence of splenomegaly, reported abdominal pain and reported diarrhoea among the different age groups. The prevalence of impaired capacity to work or study showed a strong increase with age: rising from 6.5% (2/31) among those aged 5 to 14 years to 39.7% (85/214) among those aged ⩾ 60 years. Likewise, the overall prevalence of hepatic fibrosis increased with age: rising from 16.1% (5/31) among those aged 5 to 14 years to 90.2% (193/214) among those aged ⩾ 60 years. While the prevalence of grades II–III fibrosis increased with age, the rate of grade I fibrosis reached its peak among participants aged 15 to 44 years. Grade III lesions were found only in participants aged ⩾ 15 years. An inverse association with age was found for hepatomegaly; the prevalence declined from 90.3% (28/31) among those aged 5 to 14 years to 63.6% (136/214) among those aged ⩾ 60 years.

The results obtained through the EQ-5D plus questionnaire are summarized in Tables 2 and 3. More than half of all participants with chronic schistosomiasis japonica reported impairments (moderate impairment was reported by 50.0%, 705/1410; extreme impairment by 5.8%, 82/1410), with the highest prevalence in the pain or discomfort dimension (47.9%, 675/1410) and the anxiety or depression dimension (39.4%, 555/1410). Impaired cognition and usual activities are typical sequelae of chronic schistosomiasis; impairments in cognition were articulated by 20.2% (285/1410) of the participants and impairments in ability to perform usual activities were acknowledged by 14.7% (207/1410). Impaired mobility was reported by only 2.6% (36/1410) of the population sample. Stratification of the questionnaire results by age and probing with the Cochran–Mantel–Haenszel χ² test revealed a significant linear association between age and each of the six dimensions (P < 0.001). Usually, the highest level of impairment was found in participants aged ⩾ 60 years. The youngest participants reported no impairment with regard to mobility, self-care and usual activities.

The mean disability weight of chronic schistosomiasis japonica was computed after 43 participants were excluded because they had been diagnosed with additional conditions. The overall disability weight was 0.191, and the age-specific values were 0.095 for those aged 5 to 14 years, 0.159 for those aged 15 to 44 years, 0.207 for those aged 45 to 59 years and 0.246 for those aged ⩾ 60 years. The difference between the age-specific disability weights was highly significant (ANOVA F = 61.1, P < 0.001). All pair-wise comparisons among age groups showed statistical significance at a level of 5% (Table 4).

Results obtained from the two multilevel models revealed highly significant covariation for disability weights across villages (P < 0.001); this implies clustering and variation in the degree of disability at the community level. The fixed effect of the county had no statistical significance (P > 0.05). At the individual level, the disability weight for males was significantly lower than that for females (P < 0.01). There was no significant difference with regard to educational attainment, except among the oldest age group in model 2.

The fixed and random parameter estimates are summarized in Tables 5 and 6 (available at: http://www.who.int/bulletin). In model 1, the dimensions of mobility and self-care were removed (P > 0.05). The remaining four dimensions of the questionnaire were all positively associated with the disability weight (P < 0.001). The highest disability weights were found among farmers, fishermen and boatmen. In model 2, splenomegaly and the number of previous antischistosomal treatments failed to predict the disability weight, and hence were removed from the analysis. The disability weight was significantly associated with abdominal pain, diarrhoea, blood in the stool, impaired capacity for work or study, hepatomegaly and hepatic fibrosis (P < 0.05). Grades II–III hepatic fibrosis, impaired capacity for work or study, hepatomegaly and blood in the stool accounted for most of the variation in the disability weight.

Discussion

The national schistosomiasis control programme in China, launched in the mid-1950s, is considered one of the greatest public health achievements in the country.13 However, the distribution and severity of schistosome-related morbidity at the community level remain to be fully understood. We attempted to assess the age-specific disability weight of chronic schistosomiasis japonica using a combination of tools, both objective (pathology assessed by ultrasonography) and subjective (self-reported signs and symptoms and quality of life assessed by questionnaire), in a community-based study.

A fundamental issue in the study of schistosomiasis is that infection and morbidity are not one and the same. The morbidity is the result of immunological reaction against parasite eggs that are trapped in the tissues of the human host. Detection of S. japonicum eggs in stool samples remains the gold standard for diagnosis, thus putting the emphasis on infection.1,3 This approach, however, lacks sensitivity in areas where multiple rounds of mass drug administration have been implemented.3,32,33 On the other hand, estimates of disease burden and also control programmes make reference to actual morbidity.7,34,35 The development of inexpensive serological tests with a high sensitivity has facilitated large-scale screening for schistosomiasis,33 and abdominal ultrasonography provides a convenient means for assessing hepatosplenic pathology and measuring disease progression. Hepatic fibrosis, especially grade II and grade III, is a sound marker for chronic schistosomiasis japonica.36,37 In this study, seropositive individuals with a history of recent contact with cercariae-infested water were examined by ultrasonography, and those with typical pathological features were defined as cases with chronic schistosomiasis. Cases concurrently diagnosed with other diseases (for example, hepatitis B and C) were excluded to avoid confounding when estimating the disability weight attributable to chronic schistosomiasis japonica.

Self-reports of abdominal pain, diarrhoea and blood in the stool showed promise for use in rapid diagnosis of S. japonicum infection in preceding questionnaire-based surveys.38,39 We found that the number of chronic cases and the prevalence of reported symptoms were low in the youngest age group (5 to 14 years). In general, the prevalence of self-reported signs and symptoms increased with age. Similarly, the level and extent of hepatic fibrosis increased with age, but it lagged behind that of self-reported signs and symptoms. These observations mirror the chronic nature of schistosomiasis and emphasize the importance of preventing reinfection and providing antifibrosis treatment after parasitological cure and the subsequent regression of other conditions, such as hepatomegaly.40,41 Comparisons of questionnaire results between different age groups revealed that older people (aged ⩾ 45 years) rated their quality of life as considerably poorer than their younger counterparts (those aged < 45 years). This finding corroborates the notion that older people suffer disproportionately from chronic schistosomiasis but other psychosocial factors may also play a role.

The age-specific disability weight of chronic schistosomiasis japonica rose from 0.095 in children aged 5 to 14 years to 0.246 in the oldest age group. To our knowledge, this is the first attempt to estimate the age-specific disability weights of this disease within a community-based study. The resulting values are several times higher than the weights put forward in the Global Burden of Disease study for quantifying the global burden of schistosomiasis (that is, 0.005 to 0.006). However, the latter estimates are weighted averages of all possible nonfatal health outcomes associated with a schistosome infection regardless of the species and the duration and intensity of infection, and they are based on expert opinion. Our results are closer to findings from a meta-analysis that reported disability weights of 0.02 to 0.15.9 In addition, our findings are in agreement with a report by Finkelstein and colleagues.42 Using a decision-tree model they found an overall disability weight of 0.130 for chronic schistosomiasis japonica, with age-specific disability weights of 0.098 for those aged < 15 years and 0.198 for those aged ⩾ 15 years.

The multilevel regression models indicate that disability weights are significantly associated with schistosome-related morbidity, such as hepatic fibrosis and hepatomegaly, as well as self-reported abdominal pain, blood in the stool, impairments in the capacity to work or study and in cognition.9 We found a strong heterogeneity across villages but no significant difference between the two counties. Therefore, it appears that ecological conditions govern the epidemiology of schistosomiasis at the village level and lead to clustering of disease and disability, annihilating the effects of socioeconomic conditions, which are generally thought to be an important determinant of health.5,19,43 At the individual level, age showed a positive linear association with the estimated disability weight and with many of the self-reported signs and symptoms and pathology in the univariate analysis. However, age seemed to play a less important part in the multilevel models. Thus, age was rather a confounder since it was associated both with disability and several of the health outcomes.44,45 Interestingly, the disability weight differed by sex. There is a need to study the root causes of this gender difference. We speculate that there are risk factors, adaptation mechanisms and coping strategies that are gender-specific.22,23 Farmers tended to have an elevated disability weight, which is probably due to occupation-specific risk factors.

There are at least three shortcomings of this investigation. First, because our disability estimates excluded the effects of some morbid sequelae associated with schistosomiasis (for example, anaemia and malnutrition), our disability weights are likely to yield underestimates of the true burden posed by this disease. Second, it is conceivable that some participants suffered from co-morbidity that remained undiscovered by our diagnostic approach. Third, the general health condition of people is likely to deteriorate as they age. However, we are confident that the combined effects of these limitations are rather small and, hence, the methods and calculations used resulted in reliable estimates of the age-specific disability weight of chronic schistosomiasis japonica.

The disability weight of chronic schistosomiasis japonica is considerably higher than previously estimated. Our results contribute to the further strengthening of the evidence-base of the true health impact of chronic schistosomiasis. There is a pressing need to recalculate the species-specific disability weight of chronic schistosome infections, which in turn will lead to a more accurate estimate of the global burden of this neglected tropical disease.46 This knowledge may guide a more rational and cost-effective allocation of resources for controlling schistosomiasis.3,8.9


References

Affiliations

  • National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, 207 Rui Jin Er Rd, Shanghai 200025, China.
  • Department of Public Health and Epidemiology, Swiss Tropical Institute, Basel, Switzerland.
Share