Implementing the universal health coverage: which source of information is more reliable?

The implementation of universal health coverage needs accurate data on the distribution of health benefit coverage, particularly the uninsured. The national surveys and routine reports are two important sources of information ready for use. This study shows the validation of data from two sources. The data from national household surveys on the medical welfare, the health card and the social security schemes were validated with the routine report data of the Ministry of Public Health (MOPH) and the Social Security Office (SSO) by provinces. There were considerable differences between these data sets. The national survey data gave a 1.5 times higher estimate than the report data of the MOPH and the SSO. Financial implications of using inaccurate data to implement the universal health coverage could be huge, depending on the capitation rate.

Follow up of water use in a tin mining area affected with arsenic poisoning.

Ron Phibun district in southern Thailand has been known as an endemic area for arsenic contamination. The government has been trying to improve the situation by encouraging the use of rainwater and piped water. This study aimed to document the change of water use and to identify factors associated with safe water use in 1997 compared to that in 1994. Home visits and face-to-face questionnaire interviews were undertaken. Information on water use for drinking, cooking, washing food and washing utensils in 1994 and 1997 was obtained. Among 3,849 households from which data could be obtained (estimated 79% of total households), the percentages of using safe water (including water from bottled rain water, piped and artesian well water) for drinking and cooking rose from 72.5 and 57.9 in 1994 to 93.6 and 80.9 in 1997, respectively. The percentages for washing foods and for washing utensils rose from 28.6 and 20.5 to 59.1 and 53.8, respectively. In 1997, percentage of households using piped water for drinking and cooking was still low (3.6 and 12.3) compared to those using piped water for washing food and utensils (39.1 and 43.6). Multivariate analysis shows that independent factors of the household predicting safe water use are: high arsenic area, near main road and having piped water installed. The influence of these factors (as judged by the level of odds ratio) operates more or less equally on water use for all purposes, except that installation of piped water has more influence on washing water than drinking and cooking water. We conclude that safe water supply in the area is still inadequate. Even if piped water is installed, it is often not used for drinking and cooking. The reasons for not using piped water for drinking and cooking need to be identified.

Community-based self-reported symptoms of antepartum morbidities; the health burden and care-seeking patterns of rural Bangladeshi women.

In Bangladesh there is a dearth on information relating to complications during pregnancy. We followed up 1,019 pregnant women in rural Bangladesh sampled from all the 4 old administrative divisions of the country. Trained female interviewers visited households of the pregnant women at four-week intervals and interviewed them for their current pregnancy-related complications. Out of a total of 3,812 antepartum visits the percentage of reported symptoms of bleeding, fits and convulsions, excessive vomiting, fever >3 days, urinary problems, palpitations and symptomatic anemia were 0.3, 0.7, 1.4, 4.0, 26.8, 46.5 and 78.3 respectively. Morbidities were considered to cause a health burden if they imposed constraints in daily activities of the pregnant women and they were weighted according to intensity of the constraint. For each morbidity, the mean intensity of burden per episode and the population burden per 1,000 person months of observation of all the women were calculated. For common sustaining morbidities like symptomatic anemia and urinary problems the population burden was much heavier than that for more serious but rare morbidities like bleeding and convulsions. Among the visits in which the women had any symptoms, the percentages of care-seeking for less frequently reported morbidities such as fits and convulsions, bleeding, fever >3 days, excessive vomiting were about 74, 50, 34 and 33% respectively, whereas those for more commonly reported complications such as urinary problems, symptomatic anemia and palpitations were less than 20%. Care for these morbidities was mostly sought from untrained providers.

Characteristics of repeat aborters in Vietnam.

Two hundred and sixty married women seeking induced abortion service in Hanoi, Vietnam were interviewed to determine the magnitude of repeat induced abortion and explore selected characteristics of the repeat aborters. Seventy-one percent of the sample reported having had at least one previous induced abortion. After adjustment for age and number of living children, poor attitudes toward contraception, low use of modern contraceptives and failure of contraception were shown to be significantly associated with repeat induced abortion. Woman's age, number of living children, contraceptive knowledge and experience and desire for no more children were positively related to repeat induced abortion. Socio-demographic characteristics were not related to repeat induced abortion. Improvement of attitudes toward contraception, persuasion to use modern contraception and promotion of contraceptive effectiveness are recommended strategies to prevent repeat induced abortion.

Secular increases in weight, height and body mass index among school children of Hat Yai, Thailand: a 5 years follow-up study.

Upward trend of growth and overweight has been reported from developed countries. As Thailand has been undergoing rapid economic transition over the previous decade, we conducted an analysis to demonstrate the secular changes of growth over five years follow-up. Subjects were a cohort of 2,252 primary school children recruited in 1992 for the Hat Yai childhood obesity study. Baseline demographic and family data were collected by a questionnaire completed by parents. To quantify the cohort effect, a generalized estimating equations analysis for a cross sectional time series data was undertaken with weight, height, and body mass index (BMI, kg/m2) as a dependent variable and containing a quadratic term of age, sex, year of birth and family variables. One thousand and ninety-four (48.5%) children completed 6 anthropometric measurements. Graphs of median weight, height, BMI and overweight prevalence of each birth cohort against age showed secular increases of growth and overweight, and an age effect. For each one-year younger cohort, the median weight, median height and median BMI increased by 1.22 kg, 1.25 cm and 0.23 kg/m2, respectively. An increasing trend of childhood overweight reported here may signify a need for preventing overweight and reducing weight in childhood adolescence in the future.

Soil contamination and infections by soil-transmitted helminths in an endemic village in southern Thailand.

The aim of this study was to test the association between soil contamination and infection of the household members by soil-transmitted helminths in dry and rainy seasons. A lake-side community in southern Thailand with a population of 2,340 was studied twice, in the dry season and rainy season of 1995. Fifty households were randomly selected. Soil samples near the latrine, in the yard, at the foot-washing area and under the trees were taken and analysed for presence of helminthic eggs. All members of the selected household were interviewed and stool samples obtained. Age-adjusted odds ratios of presence of Ascaris and Trichuris eggs in the household soil for ascariasis and trichuriasis were 10.5 (95% CI 1.5-77.1) and 5.5 (95% CI 2.4-12.7) in dry season and 10.4 (95% CI 2.5-43.8) and 8.3 (95% CI 3.4-20.0) in rainy season. The levels of hookworm eggs detected in the soil were too low to test the association. Soil analysis for eggs of Ascaris and Trichuris may be used to predict infections among the household members but not that for hookworm.

A survey into process and worker's characteristics in the wood furniture industry in Songkhla Province, southern region of Thailand.

A cross-sectional survey of the wood furniture industry was conducted in southern Thailand in February 1993. The aim was to examine the manufacturing process, occupational hazards at the workplace, workers' demographic characteristics, period of employment, incidence rate of work related injury and some reproductive history of workers. Altogether 69 managers and 1,000 workers participated in the study. There are 2 main types of wood industry, rubberwood and hardwood. The rubberwood industry is semi-automated with advanced technology, has a female-dominated workforce of 200-300 workers per factory and overseas-market orientation. The hardwood industry is based in small-scale workplaces ranging from 20 to 60 workers, domestic-market orientation and has a male-dominated workforce. Most of the workers were young, single, of low education and were high turnover rate laborforce, with arduous work and long working hours per week. Solvent was the most frequent chemical exposure. The person-year incidence of chemical exposure in female workers was higher than in male workers for every group of chemicals. The incidence of accidents was twice as high as the official rate. The standardized fertility ratio of female wood workers was only 51.6% of that of the Thai female population. There was a high abortion rate among women who became pregnant inside the wood industry compared to that among pregnancies outside the wood factory. Wood industry workers were exposed to occupational hazards and accident-prone work conditions.

Contamination of soil with parasite eggs and oocysts in southern Thailand.

Soil contamination with parasite eggs and oocysts was surveyed in southern Thailand in December 1994 and September 1995. The survey areas were Hat Kai Tao Village in Phatthalung Province and a slum area in a city of Songkhla Province. We used a modification of the centrifugal floatation technique with sucrose solution (specific gravity, 1.200) to recover helminth eggs and protozoa oocysts. Overall, 10 genera and 11 species of parasite eggs and oocysts were recovered. They included eight species of nematoda eggs, one species of cestoda eggs, and two species of protozoan oocysts. The definitive hosts of these parasites are dogs, cats, sheep, lizards, humans, etc. The contamination rates in two areas at different occasions varied from 55% to 72% with an average of 64%. Contamination rates of Hat Kai Tao Village in beginning and end of the rainy seasons were 72%, and 55%, respectively. Although the contamination rate in the end of rainy season was higher than that in beginning of the rainy season, the difference was not statistically significant (p > 0.05). Of the 11 parasites recovered, six were infective to humans, of which, Trichuris trichura, Ascaris lumbricoides, and Toxocara species were predominant. The mean numbers of these eggs recovered in each test were more than 10. Therefore, the contamination was estimated to be more than an egg per gram of soil sample based on the recovery efficiency (40%) of this test. These results suggested that the soil in some parts of southern Thailand may be heavily contaminated by both animal and human feces.

A case-control study of acute diarrheal disease among school-age children in southern Thailand.

We conducted a case-control study of school-age children in Phatthalung, a province in southern Thailand using a questionnaire to investigate associations of children's hygiene-related behavior and hygienic conditions in their homes with acute diarrheal disease. We compared 69 acute diarrhea (less than 7 days duration) cases that attended two hospitals in Phatthalung during August 1995 to June 1996 with 69 age-, sex- and address-matched controls in primary schools who had not suffered from diarrheal disease for the past one year before August 1995. Three factors were found to be significantly associated with acute diarrheal disease: farmer or gum planter as the occupation of father [Odds ratio (OR) 6.6; 95% confidence interval (CI) 1.7-26.1, p < 0.01], installation of a refrigerator in children's homes (OR 0.2; CI 0.1-0.8, p < 0.05), and drinking untreated water (OR 2.3; CI 0.9-6.1, p < 0.1). There was no significant difference for sources of drinking water between cases and controls. Considering the data on drinking water, the results indicated that there are some problems with quality of sources of drinking water. The results also suggested that having a refrigerator could have preventive effects on acute diarrheal disease, while inadequate behavior and unhygienic environment in the homes of farmers and gum planters might be related to acute diarrheal among school-age children.

Impact of prenatal care on perinatal mortality.

This study aimed to estimate the perinatal mortality rate (PMR) in Ho Chi Minh City, Vietnam, and to assess the association between perinatal death (PD) and prenatal care (PC). During January 1st to December 31st, 1992, among 4,809 births ascertained from 22 communes randomly selected in two areas, urban and rural, of Ho Chi Minh City, 48 still births and 66 early neonatal deaths were recorded, resulting in a weighted PMR of 25 (95% confidence interval: 20-29) per thousand. Major causes of PD were prematurity 33%, congenital malformation 15%, perinatal asphyxia 12%, perinatal infection 11%, birth injury 4%, others 8% and unknown 16%. From the whole sample, a nested case-control study was conducted on 103 cases (all the mothers whose infants died) and 309 controls (selected among the mothers of surviving infants) to assess the relationship between PD and prenatal care (PC). During unconditional logistic regression, starting prenatal care (PC) within the 1st trimester was associated with a lower risk of PD, giving an odds ratio of 0.11 (95% CI: 0.02-0.61), whereas having 4 to 8 visits gave an odds ration of 0.15 (95% CI: 0.03-0.67). It is concluded that early onset of prenatal care and having four to eight visits provide significant protection against perinatal mortality.

A study of the prevalence of and risk factors for ear diseases and hearing loss in primary school children in Hat Yai, Thailand.

2,184 primary school children in Hat Yai Municipality, Southern Thailand were screened for ear diseases and hearing loss using screening audiometry, portable tympanometry and otoscopy to identify children with possible ear problems, 754 children failed the initial screen, 728 were retested and a group of 243 were referred for ENT examination and standard audiometry-tympanometry. Children with hearing loss and otitis media were thus identified. A questionnaire to identify possible risk factors was administered. The prevalence of otitis media in this population was found to be 3.25 per cent (AOM 0.69%, OME 1.14%, COM 1.74%). No risk factor identified as significant were found. Non-significant factors included sex, age, URI last month, father smoking, mother smoking and method of transport to school as an indication of social status. This study contrasts with a previous study in 1990 which showed a 3.5 per cent prevalence of COM.

Impacts of training of village health volunteers in reduction of morbidity from acute respiratory infections in childhood in southern Thailand.

In 1990, the Ministry of Public Health of Thailand started a five-year education program on management of cases with acute respiratory infection (ARI). The objective of this-study was to test whether such a program could reduce the average number of sick days of the target children. 30 villages in the study district were randomly allocated into 15 study and 15 control villages. A 2-day training workshop for village health volunteers from the study villages was conducted. The cohort of children age below 5 years in the two areas were followed-up for 19 week in the peak season of the disease. Among the 664 and 649 target children with 67,083 and 67,984 child-days observed in the study and the control villages, 71 and 41 children, respectively, were free from any episode. The preventive odds ratio of treatment adjusted for age and village effects = 0.88, 95% (CI 0.4-1.95). The median of the average sick periods in the individuals were 27 and 34 days, respectively. After adjusting for age, episodes/month and locality, the ratio of average sick days between children in the study and control villages was 0.89 (95% CI 0.76-1.05) or 11% shorter duration without statistical significance. The adjusted odds ratios of visiting the health center, private clinics, the community hospital and provincial hospital were 0.95 (95% CI 0.69-1.25), 1.43 (95% CI 0.98-2.11), 2.01 (95% CI 1.06-3.82) and 4.71 (95% CI 1.50-14.8), respectively. The training program thus had rather little impact on morbidity of the disease but tended to promote utilization of higher level of health services by the affected children.

Evaluation of hookworm control program in southern Thailand.

An intensive hookworm control program providing the rural population of southern Thailand with mass treatment with 300 mg Mebendazole and health education commenced in 1989 with a yearly budget of over US$ 1.2 million. The current research aimed to evaluate the coverage of health education and mass treatment, the effectiveness of mass treatment and the administrative process of the control program, two years after its commencement. This evaluation consisted of three studies. In the first study, 120 villages were randomly selected, from which 840 school children, 843 adults aged between 15-59 and 844 subjects aged over 60 years were interviewed to assess awareness of hookworm, latrine availability, receipt of and compliance with antihelminthics provided, and self-treatment. In the second study, 8 villages from each of the 14 provinces were randomly chosen. Stool samples from randomly selected 4,434 subjects were examined. The third study consisted of interviewing key health planners in Bangkok and 71 health officers/workers from 32 samples villages of 8 sampled provinces. Latrine availability was 80%. Percentages of subjects who had ever heard of hookworm ranged from 70 to 95 in school children, 55 to 80 in adults and 25 to 58 in the older age group. While 63 to 85% had latrines in their homes, 27% to 47% admitted defecation outside in the past month. Percentages of coverage of drug treatment ranged from 82 to 91 in school children, 68 to 80 in adults and 54 to 59 in the older age group.(ABSTRACT TRUNCATED AT 250 WORDS)

Factor influencing case-fatality rate of septicemic children.

One hundred and fourty episodes of septicemic children seen at Songklanagarind Hospital during a period of two years were reviewed to determine factors related to mortality. One hundred episodes (70.4%) of septicemia were hospital-acquired in which 98 (69%) were caused by gram negative bacilli, with Pseudomonas being the most common agent. Thirty episodes were from gram-positive cocci and eight were from polymicrobial infections. The remainder were caused by Candida species. The overall case-fatality rate was 28.6%. Using a logistic regression model, appropriateness of antibiotic use and host status (being newborn or a compromised host compared to a normal host), but not neutropenia and acquisition of infection were the only statistically significant risk factors. The exact odds ratio of inappropriate use of antibiotic adjusted for host status was 13.6 (95% confidence limits = 5.7-32.3). Percentages of inappropriate usage among the premature and full term newborn, compromised host and normal were 50, 11, 28 and 24, respectively. Inappropriate antibiotic use was the major cause of case-fatality in the study population. It was more common among the premature newborn and compromised hosts.

Effects of piped water supply on the incidence of diarrheal diseases in children in southern Thailand.

Simple piped water supply (PWS) is increasingly popular in rural areas of southern Thailand where diarrheal diseases are quite common. The current study was carried out in 1990 in the southern part of Thailand to determine whether the use of PWS could significantly reduce the incidence rate of diarrheal diseases among the children under two years of age. A dynamic cohort study started in January 1990 in seven Muslim villages where PWS was installed for some time and there was a mixture of PWS users and non-users in each village. The cohorts comprised 126 and 137 children using and not using PWS, respectively. They were followed up weekly for one year. The crude incidence rates of diarrhea episodes for children using and not using PWS were 2.54 (SE = 0.21) and 3.52 (SE = 0.18) episodes per child-year, respectively (p < 0.01). The number of diarrheal attacks in each week was modeled as a Poisson variate using a generalized estimating equation, adjusting for correlation within the child. The analysis was carried out for each village separately and meta-analysis was used to combine these analyses. The rate ratio of PWS, adjusted for concurrent age and season, food supplement and common cold, was 0.74 (95% CI = 0.59-0.93), indicating a statistically significant reduction of one quarter. However, the proportion of users decreased with time due to management failure. It was concluded that diarrheal disease in the study area was mild but the incidence rate was high. The effect of PWS on the incidence rate of diarrhea varied with locality.(ABSTRACT TRUNCATED AT 250 WORDS)

Evaluation of village piped water: a case study in southern Thailand.

This study was conducted in a district of southern Thailand to evaluate village piped water (VPW) systems with respect to the process of planning and operation. Of 67 villages in the district, 31 were installed with one or more VPW systems using water obtained from artesian wells and distributed through the pipe with neither prior filtering nor chlorination. Seven systems were found to have stopped functioning for more than 6 months. The direct causes of failure were lack of participation from the villagers, mechanical problems and serious conflicts among the leaders. Of the running systems, only 4 were operating with meters for individual households. The charge rates were probably too low to cope with sustainable maintenance costs. The water had rust, unpleasant odors, slightly salty taste and oily surface in 50, 33, 29 and 4% of the systems respectively. Among VPW users, 8% drank water from shallow wells and 77% drank unboiled water. It was concluded that VPW in the study area was not effective for several reason. Prior assessment of management feasibility and proper community education were lacking. The running costs were probable too high, not well recognized and not covered. This led to failure in gaining participation from the villagers, which eventually led to failure or potential failure of the system and waste of capital investment. These pitfalls should be prevented prior to installation of any VPW system in the future.

A survey of neonatal tetanus and perinatal mortality in southern Muslim communities in Thailand.

The objective was to obtain the incidence rate of neonatal tetanus, perinatal mortality rate, and the coverage of health care utilization in the rural Muslim in the lower part of southern Thailand. In a retrospective field study, the rural populations of Narathiwat, Yala and Pattani provinces were studied by cluster sampling technique. The still birth rate and early neonatal mortality rate were 15.3 per 1,000 births and 11.2 per 1,000 live births, respectively. The maternal mortality rate was 1.7 per 1,000 live births. 55% of all pregnancies received 2 injections of tetanus toxoid, 80% of deliveries were performed by traditional birth attendants. The incidence rate of neonatal tetanus was 0.9 per 1,000 live births. The coverage of health service for prevention of perinatal mortality was low and these perinatal and maternal mortality rates were unacceptably high. In contrast, the incidence of neonatal tetanus was not particularly high for unknown reasons.

Transitional society, health status and international migration of Muslim villagers in the lower part of southern Thailand.

A survey was carried out in a Malay-speaking Muslim community in southern Thailand to obtain baseline data for planning of long term multidisciplinary research and development. By using a 30-cluster sampling technique, 210 households of 1,308 subjects were studied in the post-Ramadan period. It was found that the community was in a social transition. The crude birth rate was 4% and 37.6% of the households had at least one migrant. About half of these migrants had been to Malaysia and mainly worked in rubber plantations. Ninety-five per cent of the households had electricity whereas only 23.8% had a latrine. Boiled or rain water was regularly drunk in only 13.3 per cent of the households. Home-grown agricultural products were not sufficient to provide adequate food. Twenty-six per cent of the adults were unemployed and 24.6% were illiterate. Of the pregnancies 26.7% had no antenatal care and complete tetanus toxoid was given to only 27.8%. Traditional birth attendants conducted 81.1% of the deliveries and only 28.9 and 24.4% of the umbilical cords were correctly cut and correctly dressed, respectively. Breast feeding was still a common (87.8%) practice. However, complete immunization was given to only 10.8%, and 37.8% of the infants had at least one diarrheal episode in the previous month. It was concluded that high birth rate, high migration, low education, low income and bad health of infants are major problems. These problems were interlinked and needed a special multidisciplinary approach. In addition to common obstacles for routine health delivery, migration may create international complications, particularly related to maternal and child care.

A simplified financial cost-effectiveness analysis of programs for prevention of hepatitis B accidental inoculation among hospital personnel in Thailand.

There are four main strategies for preventing medical personnel from hepatitis B infection following accidental inoculation. These are: pre-exposure vaccination with and without screening for anti-HBc, post-exposure immunization with and without prior screening and with or without additional vaccination. The average cost for each strategy depends on the prevalence of exposed personnel, the cost and sensitivity of the diagnostic test, and the cumulative risk. In Thailand, if the price of vaccination decreases to US$ 10 or lower, vaccination without screening test will cost least. At present, if the cumulative risk is lower than 1 in 28, the cheapest strategy will be post-exposure immunization alone, even with vaccination. If the cumulative risk is higher than 1 in 2.5, screening test followed by vaccination with anti-HBc will be the most cost-effective.

Prevalences of some cardiovascular risk factors in a rural community in southern Thailand.

Cardiovascular disease is a growing threat in Thailand, and the prevalences of known risk factors for this disease have never been determined for the rural population. In the present study, 244 men 35-64 years of age chosen by random sampling from 11 villages of Amphoe Rattaphum, Songkhla Province, Southern Thailand were questioned to ascertain socioeconomic status, history of hypertension and of smoking. Height, weight and blood pressure were measured. Most respondents (91.1%) were rice-farmers or rubber planters with low to middle income levels. Sixty-eight per cent of respondents had never been examined for blood pressure. Of the 12 subjects previously diagnosed as hypertensive, only 1 was taking medication. Seventy-seven per cent of responders had currently been smoking an average of 13.4 cigarettes daily for an average of 27.5 years. Half of the non-smokers were ex-smokers. Only 22 subjects (9%) were above average body weight for their height. None of the subjects were obese (above 120% average body weight for height). The prevalence of hypertension (BP equal to or greater than 160/95 mmHg) and borderline hypertension (BP between 140/90 and 160/95 mmHg) was 3.7 and 9.8 per cent respectively. Eight-four per cent of the hypertensive and borderline subjects were previously unaware of their condition. We concluded that in the studied area, blood pressure screening and control is inadequate, and cigarette smoking is endemic, indicating a need for remedial public health measures.