Optimizing the malaria data recording system through a study of case detection and treatment in Sri Lanka

The potential of using malaria incidence data routinely collected from endemic regions for disease control and research has increased with the availability of advanced computer-based technologies, but will depend on the quality of the data itself. We report here an investigation into the relevance of malaria statistics provided by the routine data collection system in Moneragala, a rural malaria-endemic region in Sri Lanka. All patients (n = 321) treated for malaria in 2 clusters of health care centres (HCCs) of both the private and the public sector in the administrative regions of Moneragala and Buttala Divisional Secretariat (D.S.). Divisions were studied in December 1995/January 1996. The catchment area of these HCCs included a population resident in 53 Grama Niladhari (GN) areas, the smallest administrative units of the country. Almost equal numbers of malaria patients were detected and treated at Government and private health care institutions, and in 70% of them treatment was based on a diagnosis confirmed by microscopy. The routine data recording system, however, included only statistics from the Government sector, and only of patients whose diagnosis was microscopically confirmed. In compiling data, the origin of a case of malaria is attributed to the D.S. Division in which the institution (at which the patient was treated) was located, rather than the area in which the patient was resident, which was inaccurate because 90% of malaria patients sought health care at institutions located closest to their residence, thus crossing administrative boundaries. It also led to a loss of resolution of spatial data because patients' addresses recorded at the Government HCCs to the village-level are replaced in the statistics by the D.S. Division, which is a coarse spatial unit. Modifications to the system for malaria case recording needed to correct these anomalies are defined here. If implemented, these could result in major improvements to the quality of data, a valuable resource for the future of malaria control. The paper reiterates the call for the use of a standard spatial unit within a country to facilitate exchange of data among health and other sectors for the control of tropical diseases.     

Risk factors for acute pesticide poisoning in Sri Lanka

This report describes the characteristics of patients with acute pesticide poisoning in a rural area of Sri Lanka and, for intentional self-poisoning cases, explores the relative importance of the different determinants. Data were collected for 239 acute pesticide-poisoning cases, which were admitted to two rural hospitals in Sri Lanka. Sociodemographic characteristics, negative life events and agricultural practices of the intentional self-poisoning cases were compared with a control group. Most cases occurred among young adults and the large majority (84%) was because of intentional self-poisoning. Case fatality was 18% with extremely high case fatality for poisoning with the insecticide endosulfan and the herbicide paraquat. Cases were generally younger than controls, of lower educational status and were more often unemployed. No agricultural risk factors were found but a family history of pesticide poisoning and having ended an emotional relationship in the past year was clearly associated with intentional self-poisoning. The presence of mental disorders could only be assessed for a subsample of the cases and controls and this showed that alcohol dependence was a risk factor. This study shows that acute pesticide poisoning in Sri Lanka is determined by a combination of sociodemographic and psychological factors. Suggestions are given for interventions that could control the morbidity and mortality due to acute pesticide poisoning in developing countries.     

Control of malaria vectors with the insect growth regulator pyriproxyfen in a gem-mining area in Sri Lanka

The study was conducted in eight adjacent villages in central Sri Lanka where there are many shallow pits dug by gem miners that fill with water. These become breeding places of the main malarial vector Anopheles culicifacies, and of the second most important vector Anopheles subpictus, but not of Anopheles varuna, the third most important vector. With the help of local volunteers, data on the adult populations of these three species was collected by various standard methods, and data on the incidence of malaria cases was collected by two clinics set up for the project and through the existing hospitals. Prevalence of malaria infection in symptom-less people was investigated by mass blood surveys. On the basis of a year's pre-intervention data the villages were stratified into four with high levels of malaria transmission and four with lower transmission. Within each stratum two villages were randomly assigned for mosquito control by treating all the gem pits, as well as river bed pools, with a granular formulation of the insect growth regulator pyriproxyfen at a target dose of 0.01 mg a.i./litre. The intervention caused significant reductions in the adult populations of An. culicifacies and An. subpictus. Similarly, incidence of malaria was reduced in the intervention villages to about 24% (95% c.l. 20–29%) of that in the controls. Prevalence of parasitaemia also declined significantly. It is concluded that in this situation where, with active community participation, the breeding sites of the main vectors could be located; vector control by a highly active and persistent insect growth regulator can be a very effective means of malaria control.     

Human rabies focusing on dog ecology—A challenge to public health in Sri Lanka

Sri Lanka is among the top ten countries in the world that report the highest rate of human rabies deaths (2.8 per 1,000,000 in 2007) and animal bites requiring anti-rabies post-exposure treatment (PET) (755 per 100,000 in 2003). Dogs are the main reservoir and transmitters of rabies in Sri Lanka. Present study evaluates the effectiveness of dog rabies control strategies on reducing incidence of human rabies deaths. Analysis is based on data from last three decades and showed strong correlations between the interventions and human rabies incidence. GIS maps provided a method for illustrating the district distribution of human rabies deaths and dog population density and for recognizing districts at risk. Interrupting the natural transmission cycle of rabies in dogs would be a logical approach in eliminating dog rabies in Sri Lanka. However, interventions implemented so far, such as dog vaccination, elimination of stray dogs (abandoned in 2005), and animal birth control have been inadequate to do so. Better understanding of the ecology of stray and owned dogs (e.g. population density, population structure, confinement status) in the context of the human environment and culture, are needed to strategize the control activities, requiring coordination among regional Public Health and Veterinary services.     

Laboratory and field comparisons of pyriproxyfen, polystyrene beads and other larvicidal methods against malaria vectors in Sri Lanka

Hand-dug gem pits are important breeding sites for larvae of malaria vectors in Sri Lanka. Therefore, studies were carried out to help to select an effective, economic and convenient method that could be used to control malaria vector mosquito breeding in gem pits in a mining area. The effectiveness of four types of floating layers of polystyrene was compared in the laboratory and it was found that 2 mm expanded beads were the most effective for suffocating Anopheles larvae and pupae. The insect growth regulator, pyriproxyfen at dosages of 0.01 and 0.1 mg/l were tested in the laboratory and complete inhibition of emergence was found at both concentrations. A small-scale field trial was carried out for over a year to assess the efficacy of two concentrations of pyriproxyfen, 2 mm diameter expanded polystyrene beads, temephos, used engine oil and filling pits with soil. Pyriproxyfen only required re-application twice a year, whereas temephos or oil require 12 applications per year. Due to re-excavation by gem miners, polystyrene beads and filling of pits were not as permanent solutions as was expected. Calculations based on all available data showed that two annual treatments with pyriproxyfen at 0.01 mg/l would be the most cost-effective method with oil only slightly more expensive. However, the reduced required frequency for visiting every pit made the pyriproxyfen method the one of choice. The same low concentration of pyriproxyfen also effectively inhibited emergence of adults from river-bed pools.     

Novel ellipsoid spatial analysis for determining malaria risk at the village level

The distribution patterns of malaria incidence at a village level in Thailand were demonstrated with the use of a geographical information system (GIS), and provided the study of the malaria situation at a household level. Mosaic imageries from aerial photographs were used to create maps that contained X and Y coordinates. These digitized base maps were kept as computerized files. Standard Distance Ellipse (SDE) was used to measure the prevalence of dispersion around the mean center of malaria cases and points. Households in the SDE were at greater risk of malaria infection than those located outside the SDE. The spatial pattern of malaria incidence was investigated using spatial autocorrelation using Geary's ratio and Moran's index. Five of seven villages had a clustered spatial distribution of malaria incidence, the vector point of which had a 2-3 km range from the patient's houses. Only one village had a significant clustered spatial distribution of malaria incidence (p < 0.05). Control efforts should be focused on high-risk areas, especially those households with the heaviest caseloads. This approach would probably be more cost effective than the conventional malaria control methods. This SDE analytical technique would be a novel and useful epidemiological control method for use by public health administrators. The ellipsoidal areas required malaria control intervention.     

Thalassemia Treatment and Prevention in Uva Province,Sri Lanka: A Public Opinion Survey

Due to its excessive cost thalassemia management is a major health care problem in Sri Lanka. The majority of doctors are using only desferrioxamine (DFO), in grossly inadequate doses mainly because of its unavailability. Deferiprone (L1), which is more affordable, is not used due to fear of toxicity, as previously reported. Arthropathy attributed to L1 has been observed in some patients, and has led to the discontinuation of the drug in all patients, without scientific rationale. The proposed thalassemia prevention project for Uva Province is based on prevention of marriages between carriers. This could be achieved by carrier screening and counseling of teenagers and adolescents well before they select their partners.

In Sri Lanka, people find their marriage partners at their work place or universities, by themselves, or with the help of professional marriage brokers (they are called Kapuwa), through relatives and close friends. This process of finding a partner may also be helped by paper advertisements. However, in addition to the appearance and attitude of the prospective partner, the caste, social background and horoscope are major considerations in selecting a partner. Even when they select partners on their own at the work place or university, they keep these factors in the back of their minds to ensure social acceptance. Many relationships are given up due to objections and advice from parents when the caste or social background does not match. A horoscope is a written document that almost every child gets, written by a professional horoscope reader and depending on the time of birth. It is believed, according to the horoscope, that a person's attitudes, desires, future prospects of finding a suitable partner, could be predicted. It is rare to proceed with a marriage if the horoscope does not match. These customs are considered less seriously among educated people when they find their partner at the work place or university.

The concept of thalassemia risk-free marriages advocates promotion of marriages where at least one partner is a non-carrier. Success of such a project could be monitored at the time of marriage. This opinion survey indicates that the public is motivated to promote carrier screening and the prevention of thalassemia.     

Anopheline breeding in river bed pools below major dams in Sri Lanka

Anopheline mosquito larval surveys were carried out from September 2000 to August 2002 in Mahaweli and Kelani river beds, below five major dams in the wet and intermediate zones of Sri Lanka, to study the prevalence of anopheline species in these areas. In each study site, all permanent and semi-permanent pools were surveyed fortnightly by dipping at 6 dips/m(2) surface area of water. Larvae were collected in separate containers, staged and identified at their third and fourth stages. During each survey, the surface area and depth of pools were recorded and each reading was considered as an individual observation. River bed pools below the dams contained stagnant clean water with a little or no aquatic vegetation. The majority of pools were < or =1m(2) in surface area and < or =75 cm in depth. Anopheline mosquito breeding was seen throughout the year in each study site. The average percentage of pools positive for anopheline larvae, the number of larvae per 100 pools and 100 dips were 14.85%, 32.34 and 9.29, respectively. Thirteen anopheline species, including 10 potential vectors, namely, An. barbirostris, An. culicifacies, An. jamesii, An. maculatus, An. nigerrimus, An. peditaeniatus, An. subpictus, An. tessellatus, An. vagus and An. varuna were found breeding in the river bed pools.     

Prevalence and predictors of osteoporosis in an urban Sri Lankan population

Aim: To determine the prevalence of osteoporosis in both women and men and to determine the association between osteoporosis and selected risk factors among a sample of urban Sri Lankan adults. Methods: The population aged 35–64 years residing in 22 Grama Niladhari divisions, the smallest administrative units in the country, of the Medical Officer of Health area Ragama, Sri Lanka, was identified using the electoral list and stratified into three age groups (n = 3012). A random sample was obtained from each stratum and selected individuals were invited to participate in the study (n = 1100). A structured, interviewer‐administered questionnaire was used to collect information on demographic data and details of risk factors. Bone mineral density (BMD) was measured using a peripheral dual‐energy X‐ray absorptiometry (DEXA) scan in 700 subjects (64%). A BMD < ?2.5 was considered as osteoporosis. Although peripheral DEXA is less sensitive and specific compared to central DEXA it was used in this study as it is less costly and easily transportable. Results: Of the 700 subjects who consented to participate in the study, 60% (n = 421) were females. In subjects < 50 years, 9% of women and 3% of men had osteoporosis. In subjects ≥ 50 years, 27% of women and 7% of men had osteoporosis. Female sex (odds ratio [OR] 12.0), smoking (OR 5.5), age > 50 years (OR 3.1), low BMI (OR 3.1) and low level of education (OR 1.7) were positively associated with osteoporosis. Conclusions: Female sex, age, BMI, smoking and level of education were significant predictors of osteoporosis. The prevalence of osteoporosis increased in both women and men after 50 years.     

Plasmodium falciparum malaria transmission-blocking immunity under conditions of low endemicity as in Sri Lanka

Sera from acute primary Plasmodium falciparum patients in Sri Lanka were tested for the presence of antibodies against gamete antigens and for their functional effects of transmission blocking activity. Comparisons were made with corresponding data from a previous study from sera of patients from Papua New Guinea where malaria is more highly endemic. Although the prevalence of anti-gamate antibodies in the two groups were broadly similar, the prevalence of infectivity suppressive effects in the Sri Lankan sera (56%) was less than in Papua New Guinea sera (75%), suggesting that the generation of functionally effective transmission blocking antibodies requires prolonged exposure to multiple inoculations of malaria. In Papua New Guinea sera there was a good correlation between transmission blocking effects and antibody responses to Pfs 230, a known target of transmission blocking antibodies. Among the Sri Lankan sera no strong correlation was found between transmission blocking effects and the presence of antibodies to gamete surface antigens Pfs 230 nor Pfs 48/45 as detected by immunoprecipitation of radio-iodinated gamate proteins; a strong correlation was however, found between the intensity of response to gamete surface antigens by IFA and transmission blocking effects of these sera. It is possible therefore, that the antigens identified by IFA include non-protein moieties and that these may be the targets of transmission blocking antibodies in sera from acute primary infections of P. falciparum.     

The molecular basis for the thalassaemias in Sri Lanka

The beta-globin gene mutations and the alpha-globin genes of 620 patients with the phenotype of severe to moderate thalassaemia from seven centres in Sri Lanka were analysed. Twenty-four beta-globin gene mutations were identified, three accounting for 84.5% of the 1240 alleles studied: IVSI-5 (G-->C) 56.2%; IVSI-1 (G-->A) 15.2%; and haemoglobin E (codon (CD)26 GAG-->GAA) 13.1%. Three new mutations were found; a 13-bp deletion removing the last nucleotide in CD6 to CD10 inclusively, IVSI-129 (A-->C) in the consensus splice site, and a frame shift, CD55 (-A). The allele frequency of alpha+ thalassaemia was 6.5% and 1.1% for -alpha3.7 and -alpha4.2 deletions respectively. Non-deletion alpha-thalassaemia was not observed. Triplicate or quadruplicate alpha-globin genes were unusually common. In 1.5% of cases it was impossible to identify beta-thalassaemia alleles, but in Kurunegala detailed family studies led to an explanation for the severe thalassaemia phenotype in every case, including a previously unreported instance of homozygosity for a quadruplicated alpha-globin gene together with beta-thalassaemia trait. These findings have implications for the control of thalassaemia in high-frequency populations with complex ethnic histories.     

Quantitative ultrasound of bone and calcium intake in suburban males in Sri Lanka

Background: While only 30% of all hip fractures occurred in Asia in 1990, more than 50% will occur by the year 2050. We investigated the relationship between the Stiffness Index (SI), assessed with quantitative ultrasound, and calcium intake in a cross‐sectional survey of suburban males of different ages. Methods: From 496 people who were invited, 274 participated (55%). A single operator performed quantitative ultrasound measurements at the right calcaneus using Lunar Achilles. We derived the Sri Lankan T‐score values for SI. Calcium intake was measured using semiquantitative food frequency questionnaire to measure the previous 7 days intake. Results: There was gradual decrease in mean SI from the age of 30 years. Eighty percent of the men between 21–40 years had normal T‐scores. This percentage value fell to high 60s in men between 41–70 years. After 71 years, 35% had normal T‐scores and 30% had T‐scores less than –2.5. The mean calcium intake was 197 mg/day (95% CI 187–287 mg). Conclusions: This is the first population‐based study done in Sri Lanka regarding calcium intake and SI in males. Although few men had low T‐scores according to SI after 40 years, bone health of elderly (after 71 years) is at risk levels. The overall prevalence of low SI was negligible (4%) even with low calcium intake. Age is the only factor that influenced SI.     

Health related quality of life and its associated factors among community-dwelling older people in Sri Lanka: A cross-sectional study

BackgroundPopulation ageing has become a public health issue as it is associated with increased morbidity, institutionalization and death. These may directly affect health-related quality of life (HRQOL) of older people.PurposeThe aim of the study was to investigate HRQOL and its associated factors among community-dwelling older people in Kandy district, Sri Lanka.MethodThis cross-sectional survey involved 1300 older people. The Euro 5D-3L, International Physical Activity Questionnaire, body mass index, handgrip strength were used to measure HRQOL, physical activity and nutritional status of older people respectively. Factors associated with health-related quality of life were identified through complex sample logistic regression analysis.ResultsMajority of older people (81.9%) reported poor health-related quality of life. Middle old (aOR: 12.06, 95% CI: 5.76, 25.23), very old (aOR: 174.74, 95% CI: 39.74, 768.38), vegetarian diets (aOR: 2.13, 95% CI: 1.14, 3.96), under-nutrition (aOR: 3.41, 95% CI: 1.65, 7.04) and over-nutrition (aOR: 1.85, 95% CI: 1.04, 3.28) were significantly associated with poor HRQOL. Using dentures (aOR: 0.05, 95% CI: 0.28, 0.90) was found as a protective factor for poor HRQOL.ConclusionsHRQOL was poor among community-dwelling older people in Kandy district. Nutrition-related factors need to be further investigated to improve HRQOL among older people.     

Candidate gene study of susceptibility to cutaneous leishmaniasis in Sri Lanka

Objectives To investigate the association between selected single nucleotide polymorphisms (SNPs) in TNF, LTA and SLC11A1 genes and risk of endemic cutaneous leishmaniasis (CL) in Sri Lanka through a case–control disease association study. Methods An anonymized DNA resource representative of the Sri Lankan population was genotyped initially to establish baseline parameters. This was followed up by genotyping 200 patients and 200 matched controls. Published or modified PCR/RFLP methods were employed for genotyping. Results Comparison of the different ethnic groups showed the distribution of alleles of LTA +252 A>G to differ significantly in Tamils and Moors when compared with Sinhalese. The differences seen at allele level were also reflected in the haplotypes defined by these SNPs at the TNF locus. The case–control analysis did not show an association between the SNPs or the haplotypes investigated and CL. The distribution of these variant alleles in other populations, where they are positively associated with leishmaniasis, differed significantly from the Sri Lankan study cohort. Conclusions The selected polymorphisms do not predispose to CL in the Sri Lankan population. The study of extended haplotypes at these loci using a sufficiently powered sample collection would elaborate the findings of this study. In the face of an evolving disease pattern in the country with other forms of leishmaniasis now being reported, prevalence of polymorphisms predisposing to these forms calls for heightened surveillance and preparedness.