Climate change and human health: possible communicable disease consequences in the United Kingdom.

In May 1990 a 'Think Tank' was constituted in order to advise the Director of the Public Health Laboratory Service (PHLS) on the potential effects of global climatic change, and of world population increase and movements, on the epidemiology of infectious diseases internationally and in the United Kingdom, and to consider the implications for the PHLS. This article draws on the Think Tank deliberations and describes the scenario considered, and some of the major issues identified, with particular reference to the UK.

Thetford plastics fire, October 1991: the role of a preventive medical team in chemical incidents.

OBJECTIVES: To review the role of a medical team in the emergency management of a major polyvinyl chloride (PVC) fire in an urban area. METHODS: The district health authority's consultant in communicable disease control (CCDC) was requested to advise on the health impacts of a fire that consumed some 1000 tonnes of plastic, mainly PVC, over 72 hours and which emitted a large smoke plume that threatened the health of local residents and emergency workers alike, constituting one of the largest incidents the local emergency services had dealt with in recent years. A medical team was formed comprising the CCDC, a regional epidemiologist, an occupational physician, and a medical toxicologist. This paper is an account of this team's experience of advising on the medical management of the emergency without having any formally established role or previous training for the task. RESULTS: The main issues requiring the input of the medical team included: the possible products of combustion and their effects on health; the clinical management of those exposed; the alerting of local hospitals to the type of casualties to expect; the special health risks posed to emergency workers, especially the firemen; the need for evacuation of local residents; the risks of contamination of soil, water, and crops; the potential health impact of the plume; and the provision of expert and authoritative advice on the short and long term health implications to the public. Active surveillance systems, which included the local general practitioners and hospitals, were established and air monitoring instigated. The 46 casualties were restricted to emergency personnel who had inadvertently received exposure to the fire smoke: all recovered within 48 hours. Local residents were unharmed. CONCLUSION: The incident showed the need for preventive medical teams trained to fill a formal advisory and investigative role for chemical releases and fires, and which can play an integral part in emergency management.

Outcome of an exercise to notify patients treated by an obstetrician/gynaecologist infected with HIV-1.

Experience with hepatitis B suggests that the risk of HIV transmission from a health care worker infected with HIV to a patient will be greatest during major surgical procedures. The number of patients worldwide who are known to have undergone such procedures, been notified, and subsequently tested is still too small to be confident that the risk of HIV transmission in these circumstances is negligible. We describe a patient notification exercise, undertaken in the United Kingdom in 1991. Attempts were made to contact 1217 patients, in three health districts (A, B, and C), who had undergone surgical procedures performed by an obstetrician/gynaecologist who was infected with HIV. The exercise aimed to offer the patients reassurance, counselling and--if they wished--HIV testing. One thousand one hundred and forty-two patients (94%) were contacted, and all 520 who elected to be tested were negative for anti-HIV. The proportion of identified patients tested was 63% in district A, 35% in district B, and 61% in district C. Surgical procedures were classified retrospectively according to the likely risk (none, possible, or high) of exposure to the doctor's blood and, therefore, risk of HIV transmission. One hundred and ninety-five of those tested had undergone a procedure that carried a high risk of exposure; 179 had undergone a procedure thought to carry no risk. Patients in districts A and C who had undergone a procedure that carried a high risk of exposure were more likely to be tested than those who had not; 206 patients overall had undergone procedures that carried a high risk of exposure but were not subsequently tested.(ABSTRACT TRUNCATED AT 250 WORDS)

The incidence of meningococcal illness in the East Anglian Regional Health Authority: 1990-1991.

The rate at which notifications of meningococcal meningitis were reported by districts of the East Anglian Regional Health Authority to the Office of Population Censuses and Surveys (OPCS) varied between 6.8 and 28.0 cases per million resident population per year between 1987 and 1991. A study was conducted to find out whether this variation represented differences in incidence, completeness of notification, or reporting practices. One hundred and one cases of meningococcal illness with onset between 1 January 1990 and 31 December 1991 were identified retrospectively in residents of the East Anglian region (population 2.06 million). The ascertained incidence of meningococcal illness was 24.5 cases/million/year with a range between districts of 13.1 to 35.7 cases/million/year, similar to that expected from national data. Most of the variation in the rates of reporting to OPCS was explained by the practices of two consultants in communicable disease control (CCDCs), who reported all cases of which they were aware, irrespective of statutory notification. The study showed that communication to CCDCs was sometimes inadequate, and that control measures were not instituted in a small proportion of cases. The recommendations resulting from this study are, firstly, that OPCS should produce clear guidelines for notification and reporting. In the meantime proper officers should make their reporting practice explicit. Secondly, a sensitive case definition for meningococcal illness is needed for local monitoring of prophylactic coverage. Thirdly, CCDCs, microbiologists, clinicians, and environmental health officers should review arrangements for data exchange.

Surveillance of waterborne disease in England and Wales.

A pilot scheme, designed to improve the information on waterborne disease available nationally, was set up in five health regions from October 1991 to March 1992. Consultants in communicable disease control were asked to report each month on confirmed and suspected cases of waterborne disease, and microbiological and other contamination incidents. Twelve events were reported to the PHLS Communicable Disease Surveillance Centre (CDSC) in six months: five involved human illness and seven were contamination incidents. Six other events were reported to CDSC from regions that did not take part in the scheme. The total number of events reported was small and epidemiological evidence that linked disease with water consumption was often weak or absent. Nevertheless, the scheme provided valuable information on events associated with water and would prove useful if it were established nationally, linked with guidance on the investigation of incidents.

Assessment of costs to donor hospitals for organ transplantation.

In 1987, die Department of Health in the UK set up a working party to identify reasons contributing to a shortfall in donor organs. One recommendation was reimbursement to the District Health Authorities for costs incurred in providing the donor organs. The figure chosen was not to be seen as an incentive to donate organs, merely as an appropriate compensation for the costs incurred. There would be no direct payment to doctors, trustees or relatives of the donor. With the development of the competitive health care environment in the United Kingdom, the reimbursement of donating hospital costs is being considered with these data.

The relation of maternal weight to the blood pressures of Gambian children.

The objective of the study was to relate blood pressure levels in children to their mother's weight in pregnancy. The blood pressures of 675 children aged from one to nine years in three villages in rural Gambia were measured. They were matched to antenatal clinic data which had been collected from all pregnant women in the three villages since 1980. Among children under eight years of age those born in the dry season had the highest blood pressures and were heavier. Their blood pressures were positively related to body weight and to mothers' weight at six months of pregnancy. These relationships were independent of mothers' age and parity, birthweight, gestational age, and placental weight. Among older children, aged eight and nine years, those born in the rainy season had the highest blood pressures. Their blood pressures were not related to their mothers' weight at six months of pregnancy. Rather they were inversely related to mothers' weight gain in the last trimester. An interpretation of these findings is that among young children differences in blood pressure are largely determined by rates of maturation. However, the long-term effects of adverse intra-uterine influences which elevate blood pressure become apparent in older children.

PIP: Recent findings show that the intrauterine environment has a major effect on blood pressure and hypertension. This article reports on a study of the blood pressure of 675 Gambian children 1-9 years old and the relationship to mother's weight in pregnancy and seasonality. The children were born in the rural Keneba, Manduar, and Kanton Kunda in The Gambia after January 1980. Since 1979, clinic data was available on child's birth weight, growth, and morbidity data within the 1st 18 months, and after 18 months, clinic visit data on weight, height, diagnosis, and treatment. Mother's anthropometry and blood pressure (BP) were used at within 15 days of 6 months and 30 days of birth; mother's BP is not comparable between younger and older children. Periodic census data on households was also accessible. A study survey was also administered in 1989. Children's height was measured with a Harpenden infantometer for those 900 mm, and with a Raven Maximeter for those 900 mm. A SECA medal 727 (5 g gradations) was used for lower weights and a SOEHNLE digital scale for mother's and for heavier children. DINAMAP (model 18465 X) was used for 2 BP readings. A BP pretest with 2 observers revealed a mean difference in readings of .27 and a standard error of 1.36. Seasonality was set at rainy (Aug-Nov), cool (Apr-Jly), and hot and dry (Dec-Mar). The results were that systolic and diastolic pressures rose with children's age; i.e., from 89.3 mm Hg in 1 year olds to 102.7 mm Hg in 9 year olds. It increased with body weight and decreased with the time since the last meal. Ambient temperature was unrelated. Cole's LMS method for calculating centile curves was used to adjust the data for age, body weight, and time since eating. In the age groups 1-4, 5-7, mean BP was lowest during the rainy season and among 8-9 year olds, BP was highest during rainy seasons and variation was statistically significant. Mother's weight gain was lowest in the rainy season and among 8-9 year olds, BP was highest during rainy seasons and variation was statistically significant. Mother's weight gain was lowest in the rainy season and BP increased with age. Every 10mm Hg rise in mother's systolic BP was related to a 1 mm Hg rise in children's (1-7 years) adjusted systolic BP (p=.008). Mother's weight at 7.5 months was positively related to the child's (1-7 years) adjusted systolic BP (p=.003). There was no trend for children 8-9 years. Among the younger children BP was not related to mother's weight gain. Birthweight was unrelated to adjust systolic pressure at any age, but strongly positively related to mother's height, mother's weight at 7.5 months, and weight gain in the last trimester. Future followup is planned to examine whether the 6-7 year olds a year later begin to show the effects of maternal weight gain in pregnancy.

An unusual community outbreak of influenza A.

The Hong Kong H3N2 subtype of influenza A virus appeared in 1968 and since then has caused epidemics of varying degrees of severity. We describe a community outbreak of influenza A H3N2 which occurred in members of a bowls club in an English rural village in late April 1989. The explosive onset, high attack rate (34/41 = 83 per cent) in those exposed, and the clinical presentation initially suggested a toxic or allergic aetiology. Twenty-three persons consulted their general practitioners; before the diagnosis was made all cases were considered to merit antibiotic therapy and 17/23 were prescribed steroids or bronchodilators on account of persisting severe wheeze and chest tightness. One of the 23 was admitted to hospital. There were no deaths. Influenza vaccination is recommended for people at special risk but protective efficacy is relatively low and short-lived. Only two of the group had received influenza vaccination since the beginning of October 1988 and both became symptomatic. The use of amantadine for the prevention and early treatment of influenza A in selected situations merits serious consideration, provided the diagnosis is made sufficiently early.

Breast feeding, nutritional state, and child survival in rural Bangladesh.

The effect of breast feeding on nutritional state, morbidity, and child survival was examined prospectively in a community in rural Bangladesh. Every month for six months health workers inquired about breast feeding and illness and measured arm circumference in an average of 4612 children aged 12-36 months. Data from children who died within one month of a visit were compared with those from children who survived. Roughly one third of the deaths in the age range 18-36 months were attributable to absence of breast feeding. Within this age range protection conferred by breast feeding was independent of age but was evident only in severely malnourished children. In communities with a high prevalence of malnutrition breast feeding may substantially enhance child survival up to 3 years of age.

PIP: Breastfeeding enhances survival during infancy and ranks among the 1st 4 strategies promoted by UNICEF for improving infant and child survival. Nevertheless, its impact in older children is still unknown. The effect of breastfeeding on nutritional state, morbidity, and child survival was examined prospectively in a community in rural Bangladesh. Every month for 6 months health workers inquired about breastfeeding and illness and measured arm circumference in an average of 4612 children aged 12-36 months. Data from children who died within 1 month of a visit were compared with those from children who survived. Roughly 1/3 of the deaths in the age range 18-36 months were attributable to absence of breastfeeding. Within this age range protection conferred by breastfeeding was independent of age but was evident only in severely malnourished children. In communities with a high prevalence of malnutrition breastfeeding may substantially enhance child survival up to 3 years of age. Family planning may help, as pregnancy was the most common cause of stopping breastfeeding early in this community.

Impact of infection on the growth of children from 0 to 2 years in an urban West African community.

To determine the relationship between growth and morbidity in the first 2 y of life, we studied a cohort of 126 newborns in a Gambian township. Mean weight-for-age exceeded the National Center for Health Statistics (NCHS) standards in the first half of infancy but there was a mean deficit of 1.2 kg by age 1 y. Only two diseases contributed significantly to weight faltering: diarrheal diseases were estimated to cause one-half of the deficit and lower respiratory tract infections (LRTI) one-quarter. LRTI reduced weight gain in young children by 14.7 g/d of infection and diarrheal diseases in weaning infants by 14.4 g/d. Diarrhea had no significant impact on the growth of exclusively breast-fed infants. Growth velocity was normal in the second year of life, despite continuing infections.

PIP: Previous studies have showed that diarrhea was the main nondietary cause of weight faltering in children aged 6 months to 3 years. To help determine the relevance of these findings in a wider context and to investigate whether some children were more affected than others, a group of young children from a Gambian township was studied longitudinally to determine the relationship between their growth, morbidity, and feeding patterns. This paper describes the quantitative impact of various categories of disease on the growth of these children. 126 newborns in the 1st 2 years of life were studied in Bakau. Mean weight-for-age exceeded the National Center for Health Statistics standards in the 1st 1/2 of infancy but there was a mean deficit of 1.2 kg by age 1 year. Only 2 diseases contributed significantly to weight faltering: diarrheal diseases were estimated to cause 1/2 of the deficit and lower respiratory tract infections (LRTI)1/4. LRTI reduced weight gain in young children by 14.7 grams per day of infection and diarrheal diseases in weaning infants by 14.4 grams per day. Diarrhea had no significant impact on the growth of exclusively breast-fed infants. Growth velocity was normal in the 2nd year of life, despite continuing infections.

Arm circumference and other factors in children at high risk of death in rural Bangladesh.

Mid upper arm circumference (MUAC) was measured monthly for 6 months in about 500 children aged 6-36 months from rural Bangladesh. Children who would die within 1 month of screening could be identified with 94% specificity and 56% sensitivity--almost twice the sensitivity achieved by other anthropometric screening schemes for this level of specificity. Specificity was slightly improved when the absence of breast-feeding, concurrent diarrhoea, oedema, and acute respiratory infection were taken into account. Children at high risk of death can be detected by monthly measurement of MUAC, which may be used in poor communities where interventions have to be selective.

PIP: Mid upper arm circumference (MUAC) was measured monthly for 6 months in about 5000 children aged 6-36 months from rural Bangladesh. Children who would die within 1 month of screening could be identified with 94% specificity and 56% sensitivity--almost twice the sensitivity achieved by other anthropometric screening schemes for this level of specificity. Specificity was slightly improved when the absence of breastfeeding, concurrent diarrhea, edema, and acute respiratory infection were taken into account. Children at high risk of death can be detected by monthly measurement of MUAC. This information may then be used to target resources in poor communities where interventions have to be selective. Children with a MUAC of less than 110mm are obviously malnourished and in communities where severe childhood malnutritiom still occurs, screening and referral of these high risk children may be one of the more effective stategies for the prevention of death.

Riboflavin, folate and vitamin C status of Gambian women during pregnancy: a comparison between urban and rural communities.

Anthropometric, haematological and vitamin status indices were measured in a group of pregnant women living in the urban community of Bakau in The Gambia, West Africa. Their haematological and anthropometric indices were generally within normal limits. Vitamin C status was also acceptable and was similar to values observed in rural Gambian women sampled at the same time of year. Folate status was similar to that seen previously in rural Gambian communities, and there was a strong intrasubject correlation between plasma and red cell folate levels, together with a trend towards higher values as pregnancy progressed. This was consistent with probable compliance with local recommendations for folate supplementation during pregnancy. The urban Gambian women also resembled their rural counterparts in having very poor biochemical riboflavin status, which deteriorated as pregnancy progressed. In this respect they differed markedly from UK women, who had satisfactory riboflavin status even in late pregnancy. The existence of severe biochemical riboflavin deficiency, even in urban Gambian women, whose anthropometric indices are not compatible with severe general malnutrition, suggests that a deficiency of this vitamin may be widespread in Sahelian West Africa. Measures to improve maternal vitamin status during pregnancy would therefore be equally appropriate in both rural and urban communities.

Oral rehydration solution--too little or too much?

PIP: Data from studies on actual use of packaged oral rehydration solution (ORS) in Bangladesh are summarized and their impact evaluated. In 6 studies in clinic settings only 2 clinics used enough ORS to meet minimum recommendations, of 50 ml/kg for initial rehydration in mild cases and 100 ml/kg per day of continued diarrhea, plus 100 ml/kg per day for normal requirements. The consumption of ORS in home settings was reported as 1 liter pack in 73 and 77% of adults, and 87% of children; 15 and 23% received 2 packs; 3% received 3 packs. In the authors' survey infants received 0.95 liter per episode, and adults 1.2 liter per episode. In another survey the median daily consumption was 0.3 liters. The authors estimated that the maximum weight of a person given 1 liter of ORS, according to the recommended dose, would be 12.5 kg, or a child less than 2 years old, with moderate dehydration. Considering that there are an estimated 85 million episodes of diarrhea yearly in Bangladesh, and only 4-9% are associated with dehydration, it would be more efficient for the government to provide ORS packs to truly dehydrated patients, rather than to try to treat all cases of diarrhea. Treating 5 million episodes of potentially life-threatening diarrhea with an average of 4 liters, rather than many more cases with inadequate amounts as is being done currently, would cost 1/6 as much. It is necessary to define criteria for communities to distinguish life-threatening dehydration for intensive ORT treatment.^ieng

The Gambia and Bangladesh: the seasons and diarrhoea.

PIP: Climactic factors in the Gambia and Bangladesh have an important impact on the incidence of diarrheal disease. Both countries share some common characteristics in climate, including a cool dry winter of 3 months followed by a hot dry spring and hot wet summers of 5-7 months in length. The main difference is in the amount of rainfall. The Gambia may have 20-30 inches of rain each year; Bangladesh usually has up to 4-5 times this amount. In the Gambia, drought is a recurring problem; floods is the problem in Bangladesh. A study in the Gambia found a close link between the time of the annual peak in diarrhea in young children and the summer rains. A 2nd peak of diarrhea in the winter also was significant and was shown to coincide with a short period of intense transmission of rotavirus. Of the enteric infections of childhood, the enterotoxigenic "Escherichia coli" (ETEC), that is those producing heat-stable toxin (ST) were found to be the most important etiological agents of diarrhea in both countries, with a peak during the rains. In rural Gambia, water is obtained almost exclusively from surface wells, 15-20 meters deep. It was found that, although this water was fecally contaminated throughout the year, levels of contamination increased by up to 100 times with 1-2 days of the start of the rains because excreta is washed into the wells. It also was clear that contaminated water and domestic environment contribute to contamination of children's food. The high level of contamination of food during the summer coincided with the time of high diarrhea prevalence. In Bangladesh it was shown that the incidence of ETEC diarrhea in infants was positively correlated with the frequency of consumption of weaning foods contaminated with fecal coliforms. The seasonal peak of ETEC diarrhea coincided with the time when food was most contaminated due to higher bacterial growth caused by high temperatures. Cholera is endemic in many areas of Bangladesh but not in the Gambia. Though similar some other diarrheal diseases in showing a rainy season peak, the timing of peaks of cholera incidence can and has changed from year to year in Bangladesh. The reason for this and the variable occurence of a less marked pre-rains peak of cholera is unknown. During the main farming season, mothers have less time for breastfeeding. Smaller amounts of breast milk were consumed by breastfeeding infants at this time of year coinciding with poorer maternal nutritional status. This also was the season of poorest nutritional status in children, leading to increased duration and, perhaps, severity of diarrhea. Personal hygiene, attitudes toward breastfeeding, and weaning practices are important, non-seasonal factors in diarrheal diseases. To be more effective, health education messages could be varied according to the season, for different problems occur at various times of the year.^ieng

The weanling's dilemma: are we making progress?

In many developing countries the weanling child (the breastfed child who is regularly receiving additional food) still suffers a high level of morbidity and mortality from diarrhoeal disease. The initiation of weaning is a critical event. No clear strategy exists for substantially enhancing the breast milk output of demand feeding mothers in underprivileged communities. Remarkably little progress has been made in our ability to advise mothers, either on a collective or an individual basis, as to when they should supplement the diet of their breastfed offspring, one continuing problem being the failure to adopt appropriate growth standards for infants. Furthermore there has been little attempt to improve traditional weaning foods in terms of consistency, shelf life and bioavailability of nutrients. Increasing insights into the normal growth pattern of breastfed infants and knowledge of localised appropriate traditional food technology remain grossly underexploited.

The etiology of diarrhoea studied in the community in young urban Gambian children.

PIP: In a community-based longitudinal study of 126 urban Gambian children from birth to 2 years, bacterial and/or viral pathogens were identified significantly more often in diarrheal episodes studied (27.5%) than during control periods (19.0%), (p0.001). Rotavirus and ST-toxigenic Escherichia coli emerged as the most frequently-occurring pathogens in symptomatic children. Campylobacter species, accounting for nearly 1/3 of all pathogens, isolated, occurred more frequently during control periods, as did Salmonella species. Shigellosis may have been underdiagnosed. Even so, the cause of diarrhea during the 1st 1/2 of infancy was largely unresolved, with approximately 90% of cases studied being unexplained, a similar figure emerging for all attacks studied during the spring-time 1/4 of the year. We remain notably ignorant of the cause of most unselected cases of childhood diarrhea in the community in West Africa.^ieng

Epidemiological aspects of rotavirus infection in young Gambian children.

Rotavirus gastro-enteritis in young Gambian children has its maximum impact on infants after the age of one month, in whom it produces short, well-defined annual winter epidemics with clinical dehydration in up to 18% of those infected. Sporadic infection was observed in neonates who were often asymptomatic, throughout one year but not in the subsequent year. In two consecutive years studied there was a major change from subgroup I, serotype 2 to subgroup II, serotypes 1 and 3. This could have contributed to the failure of children to develop protective immunity against sequential disease following an infection during infancy. If rotavirus morbidity in this community is to be notably reduced by a vaccination programme it would need to be carried out in early infancy prior to the winter season. Evaluation of a type-specific vaccine should include monitoring secular changes in rotavirus serotypes throughout subsequent epidemics.