Differences in the risk of cervical cancer and human papillomavirus infection by education level

BACKGROUND

Cervical cancer risk is associated with low education even in an unscreened population, but it is not clear whether human papillomavirus (HPV) infection follows the same pattern.

METHODS

Two large multicentric studies (case-control studies of cervical cancer and HPV prevalence survey) including nearly 20 000 women. GP5+/GP6+ PCR was used to detect HPV.

RESULTS

Education level was consistently associated with cervical cancer risk (odds ratio (OR) for 0 and >5 years vs 1-5 years=1.50, 95% confidence interval (CI): 1.25-1.80 and 0.69, 95% CI: 0.57-0.82, respectively, P for trend <0.0001). In contrast, no association emerged between education level and HPV infection in either of the two IARC studies. A majority of the women studied had never had a Pap smear. The association between low education level and cervical cancer was most strongly attenuated by adjustment for age at first sexual intercourse and first pregnancy. Parity and screening history (but not lifetime number of sexual partners, husband's extramarital sexual relationships, and smoking) also seemed to be important confounding factors.

CONCLUSION

The excess of cervical cancer found in women with a low socio-economic status seems, therefore, not to be explained by a concomitant excess of HPV prevalence, but rather by early events in a woman's sexually active life that may modify the cancer-causing potential of HPV infection.

Different cervical cancer screening approaches in a Chinese multicentre study

To evaluate alternative cervical cancer screening methods, digital colposcopy and collection of cervical exfoliated cells for liquid-based cytology (LBC) and hybrid capture 2 (HC2) testing were performed among 2562 women aged 15–59 years in three study sites in the People's Republic of China (rural Shanxi province, Shenyang city in Liaoning province and Shenzhen city in Guangdong province). Visual inspection with acetic acid (VIA) was also evaluated independently from colposcopy. A total of 74 cases of histologically confirmed cervical intraepithelial neoplasia grade 2 or worse (CIN2+) were identified, and 16 CIN2+ cases were imputed among unbiopsied women to correct for verification bias. Corrected sensitivity for CIN2+ was 37% for VIA, 54% for colposcopy, 87% for LBC with a threshold of atypical cells of undetermined significance (LBC⩾ASCUS), 90% for HC2, 84% for LBC using HC2 to triage ASCUS and 96% for positivity to LBC⩾ASCUS or HC2. For VIA, sensitivity was much lower among women ⩾40 years (12%) than those aged ⩽39 years (50%). Specificity varied from 77% for positivity to LBC⩾ASCUS or HC2, up to 94% for LBC using HC2 to triage ASCUS. In conclusion, LBC, HC2 and their combinations performed well, whereas VIA missed a majority of CIN2+, particularly in older women. Digital colposcopy performed better than VIA, but still missed nearly half of CIN2+ in this study.

Carcinoma of the cervix and tobacco smoking: collaborative reanalysis of individual data on 13,541 women with carcinoma of the cervix and 23,017 women without carcinoma of the cervix from 23 epidemiological studies

Tobacco smoking has been classified as a cause of cervical cancer, but the effect of different patterns of smoking on risk is unclear. The International Collaboration of Epidemiological Studies of Cervical Cancer has brought together and combined individual data on 13,541 women with and 23,017 women without cervical carcinoma, from 23 epidemiological studies. Relative risks (RRs) and 95% confidence intervals (CIs) of carcinoma of the cervix in relation to tobacco smoking were calculated with stratification by study, age, sexual partners, age at first intercourse, oral contraceptive use and parity. Current smokers had a significantly increased risk of squamous cell carcinoma of the cervix compared to never smokers (RR = 1.60 (95% CI: 1.48-1.73), p<0.001). There was increased risk for past smokers also, though to a lesser extent (RR = 1.12 (1.01-1.25)), and there was no clear trend with time since stopping smoking (p-trend = 0.6). There was no association between smoking and adenocarcinoma of the cervix (RR = 0.89 (0.74-1.06) and 0.89 (0.72-1.10) for current and past smokers respectively), and the differences between the RRs for smoking and squamous cell and adenocarcinoma were statistically significant (current smoking p<0.001 and past smoking p = 0.01). In current smokers, the RR of squamous cell carcinoma increased with increasing number of cigarettes smoked per day and also with younger age at starting smoking (p<0.001 for each trend), but not with duration of smoking (p-trend = 0.3). Eight of the studies had tested women for cervical HPV-DNA, and in analyses restricted to women who tested positive, there was a significantly increased risk in current compared to never smokers for squamous cell carcinoma (RR = 1.95 (1.43-2.65)), but not for adenocarcinoma (RR = 1.06 (0.14-7.96)). In summary, smokers are at an increased risk of squamous cell but not of adenocarcinoma of the cervix. The risk of squamous cell carcinoma increases in current smokers with the number of cigarettes smoked per day and with younger age at starting smoking.

Papillomavirus infection in rural women in southern India

To investigate the prevalence of, and the risk factors for, cervical infection with 44 types of human papillomavirus (HPV) in a rural area in the Dindigul District, Tamil Nadu, India, we interviewed and obtained cervical cell samples from 1891 married women aged 16-59 years. HPV prevalence was 16.9% overall and 14.0% among women without cervical abnormalities, or 17.7 and 15.2%, respectively, age-standardised to the world standard population. In all, 21.9% of infections involved more than one HPV type. High-risk HPV types predominated, particularly HPV 16 (22.5% of women infected), followed by HPV 56, HPV 31, HPV 33 and HPV 18. Unlike most populations studied in developed countries, HPV prevalence was constant across the age groups. HPV positivity was inversely associated with education level (odds ratio (OR) among women with high school vs no education=0.6) and positively associated with widowhood and divorce (OR=1.7), nulligravidity (OR=2.3), and condom use (OR=2.6). It is unclear how much low clearance of, or frequent reinfection with HPV accounted for the study prevalence of infection in different age groups.

The sexual health of pupils in years 4 to 6 of primary schools in rural Tanzania

BACKGROUND/OBJECTIVES

There is an urgent need for effective interventions to improve the sexual and reproductive health of adolescents. Reliable data on the sexual health of adolescents are needed to guide the development of such interventions. The aim was to describe the sexual health of pupils in years 4 to 6 of 121 rural primary schools in north western Tanzania, before the implementation of an innovative sexual health intervention in 58 of the schools.

METHODS

A cross sectional survey of primary school pupils in rural Tanzania was carried out. The study population comprised pupils registered in years 4 to 6 of 121 primary schools in 20 rural communities in 1998. Basic demographic information was collected from all pupils seen. Those born before 1 January 1985 (aged approximately 14 years and over) were invited to participate in the survey, and asked about their knowledge and attitudes towards sexual health issues, and their sexual experience. A urine specimen was requested and tested for HIV, Chlamydia trachomatis (CT), Neisseria gonorrhoeae (NG) and, for females, pregnancy.

RESULTS

9283 pupils born before 1 January 1985 were enrolled and provided demographic information and a urine sample. Male pupils were significantly older than females (mean age 15.5 years v 14.8 years, p<0.001), but all other demographic characteristics were similar between the sexes. 14 (0.2%) of the enrolled pupils (four male and 10 female) were HIV positive, 83 (0.9%) were positive for CT, and 12 (0.1%) for NG. 32 female pupils (0.8%) were positive by pregnancy test. Sexual experience was reported by one fifth of primary school girls, and by almost half of boys. Only 45/114 (39%) girls with biological markers of sexual activity reported having had sex.

CONCLUSIONS

HIV, CT, NG, and pregnancy were present though at relatively low levels among pupils in years 4 to 6 of primary school. A high proportion of pupils with a biological marker of sexual activity denied ever having had sex. Alternative ways of collecting sensitive data about the sexual behaviour of school pupils should be explored.

Human papillomavirus infection and use of oral contraceptives

Human papillomavirus (HPV) infection is thought to be a necessary but not sufficient cause of most cases of cervical cancer. Since oral contraceptive use for long durations is associated with an increased risk of cervical cancer, it is important to know whether HPV infection is more common in oral contraceptive users. We present a systematic review of 19 epidemiological studies of the risk of genital HPV infection and oral contraceptive use. There was no evidence for a strong positive or negative association between HPV positivity and ever use or long duration use of oral contraceptives. The limited data available, the presence of heterogeneity between studies and the possibility of bias and confounding mean, however, that these results must be interpreted cautiously. Further studies are needed to confirm these findings and to investigate possible relations between oral contraceptive use and the persistence and detectability of cervical HPV infection.

Human papillomavirus types in invasive cervical cancer worldwide: a meta-analysis

This study investigated regional variations in the contribution made by different human papilloma (HPV) types to invasive cervical cancer (ICC). A total of 85 studies using polymerase chain reaction to estimate HPV prevalence in ICC were identified. Data on HPV prevalence were extracted separately for squamous cell carcinoma (SCC) and for adeno- and adenosquamous-carcinoma (ADC). A total of 10 058 cases (8550 SCC, 1508 ADC) were included in pooled analyses. The most common HPV types in ICC were, in order of decreasing prevalence, HPV16, 18, 45, 31, 33, 58, 52, 35, 59, 56, 6, 51, 68, 39, 82, 73, 66 and 70. In SCC, HPV16 was the predominant type (46-63%) followed by HPV18 (10-14%), 45 (2-8%), 31 (2-7%) and 33 (3-5%) in all regions except Asia, where HPV types 58 (6%) and 52 (4%) were more frequently identified. In ADC, HPV prevalence was significantly lower (76.4%) than in SCC (87.3%), and HPV18 was the predominant type in every region (37-41%), followed by 16 (26-36%) and 45 (5-7%). The overall detection of HPV DNA was similar in different regions (83-89%). A majority of ICC was associated with HPV16 or 18 in all regions, but approximately a quarter of all ICC cases were associated with one of 16 other HPV types, their distribution varying by region.

Prevalence and determinants of human papillomavirus genital infection in men

Four-hundred-forty-five husbands of women with invasive cervical carcinoma, 165 of women with in situ cervical cancer, and 717 of control women (age range 19-82 years) were interviewed and a sample of exfoliated cells from the penis obtained in seven case-control studies conducted by the International Agency for Research on Cancer. The characteristics of human papillomavirus-positive and human papillomavirus-negative husbands were compared using odds ratios and 95% confidence intervals. Thirteen per cent of the husbands of control women, 18% of the husbands of women with invasive cervical carcinoma, and 21% of the husbands of in situ cervical carcinoma women were positive for penile human papillomavirus DNA. Human papillomavirus 16 was detected in 45 husbands, human papillomavirus 18, 31 or 33 in 19, and human papillomavirus 6/11 in 6, but the majority of human papillomavirus infection (158) was with other or unspecified human papillomavirus types. The same human papillomavirus type was seldom identified in both husband and wife. The strongest variation in penile human papillomavirus infection was by country, with percentages among the husbands of control women ranging between 3% in Spain and 39% in Brazil. Having had over 50 lifetime sexual partners, compared with only one, was associated with an odds ratio of 2.3.

A case-control study of gastric cancer in Venezuela

A case-control study to evaluate risk factors for gastric cancer was carried out among 292 cases of gastric cancer and 485 controls in a high-risk area of Venezuela. Subjects were interviewed using a structured questionnaire, which elicited information on residential history, socio-economic status, family history of gastric diseases, smoking, drinking and dietary habits. Habitual diet was estimated from a meal-structured food frequency questionnaire on 75 food items. There was a strong inverse association with social class, as measured by education and by indicators of poverty. The results of the dietary analysis suggest that a diet high in starch and low in meat, fish and fresh vegetables increases risk of gastric cancer. A protective effect was observed for frequent consumption of allium vegetables. Inverse associations were found with height, which may reflect nutritional status in childhood, and with refrigerator use in the first two decades of life. Alcohol and tobacco consumption was investigated among males only, since the prevalence of alcohol and tobacco use was very low in females. Alcohol drinkers were at higher risk than non-drinkers and there was a small excess risk for current smokers compared with never smokers. There was some evidence of familial aggregation of gastric cancer. These findings will have important implications in planning preventive strategies for gastric cancer in Venezuela.

Helicobacter pylori eradication attenuates oxidative stress in human gastric mucosa

OBJECTIVE

Helicobacter pylori infection causes gastric diseases, but the responsible mechanisms are not completely understood. They can involve DNA and tissue damage induced by reactive oxygen and nitrogen species. Our aim is to investigate the effects of bacterial eradication on oxidative stress by measuring changes of relevant markers.

METHODS

Antral biopsies were obtained from 34 patients with chronic atrophic gastritis and peptic ulcer disease before and after bacterial eradication. The expression of inducible nitric oxide synthase (iNOS) and levels of nitrotyrosine (NTYR) and 8-hydroxy-2'-deoxyguanosine were assessed immunohistochemically as markers of nitric oxide (NO) production and of damage to proteins and DNA, respectively.

RESULTS

Before treatment, the percentages of patients with staining were: 56 for iNOS in inflammatory cells, 79 and 61 for NTYR and 8-hydroxy-2'-deoxyguanosine in foveolar cells, respectively, and 82 for 8-hydroxy-2'-deoxyguanosine in lymphoid follicles. NTYR staining was associated with the intensity of inflammation (p = 0.04) and gastritis activity (p = 0.07). The prevalence of 8-hydroxy-2'-deoxyguanosine tended to be associated with that of NTYR. After successful H. pylori eradication, the prevalence of iNOS and NTYR (in mild gastritis) staining decreased (p < 0.001 and p < 0.06, respectively). 8-Hydroxy-2'-deoxyguanosine staining disappeared in 24% of cases but appeared in 18% of previously negative cases despite eradication.

CONCLUSION

Targets of oxidative stress associated with H. pylori infection are inflammatory and deep foveolar cells and lymphoid follicles. This is the first report of 8-hydroxy-2'-deoxyguanosine localization in gastric mucosa. Oxidative stress is reduced by bacterial eradication in the first stages of mild gastritis. Moderate-severe gastritis may be a step that is reversible for iNOS, but partly irreversible for NTYR and 8-hydroxy-2'-deoxyguanosine.

Helicobacter pylori and stomach cancer: a case-control study in Venezuela

The role of Helicobacter pylori infection in gastric cancer was evaluated in a high-risk population in Venezuela using serological assays in a study of 302 cases and 483 neighborhood controls. To investigate the claim that assays for H. pylori should use antigens derived from local strains, four different assays derived from Venezuelan and European strains were used. Prevalence of IgG H. pylori antibodies in controls was very high, with estimates between 72 and 92%. Prevalence was similar in cases and controls. However, cases had lower antibody titers. This effect was observed only in subjects with low pepsinogen (PG) levels PGI/PGII <3.0), which suggested that extensive atrophy in cases causes a loss of H. pylori infection, with a consequent reduction in antibody titer. In addition, advanced cases (stage II or higher) had lower antibody titers than less advanced cases, which indicated that the lower antibody titers in cases may be attributable partially to a diminished immune response. All of the four assays for anti-H. pylori antibodies gave similar results. No evidence was found for the superiority of the assay based on Venezuelan strains. These results are consistent with other case-control studies in high-risk populations and highlight the difficulties of investigating H. pylori infection in retrospective studies.

Lewis antigen alterations in a population at high risk of stomach cancer

Anomalous Lewis(a) antigen and sulfomucin expression are considered as markers of progression in precursor lesions of gastric cancer. Additionally, Lewis antigen and secretor phenotype have been related to Helicobacter pylori infection and gastric epithelial damage. The two objectives of this study were to correlate Lewis antigen alterations with histochemical changes and to explore the relationship between Lewis and secretor phenotypes and gastric epithelial damage related to H. pylori infection. The study subjects were selected from a chemoprevention trial in Tachira State, Venezuela, an area with a high risk of gastric cancer. Anomalous Lewis(a) antigen expression in Lewis (a-b+) phenotype individuals was closely related to the severity of the histological lesions, especially to dysplasia and type III intestinal metaplasia lesions. A weak relationship was observed between nonsecretor individuals and more advanced lesions of IM, but this association was not statistically significant. There was no relationship between secretor phenotype and H. pylori status, atrophy, regenerative activity, erosion, or ulcer.

Population-based study of human papillomavirus infection and cervical neoplasia in rural Costa Rica

BACKGROUND

Human papillomavirus (HPV) is the main cause of cervical neoplasia. Because few population-based studies have investigated the prevalence of type-specific infection in relation to cervical disease, we studied a high-risk population, estimating the prevalence of HPV infection and the risk associated with various HPV types.

METHODS

We screened 9175 women in Guanacaste, Costa Rica, to obtain a referent standard final diagnosis, and tested 3024 women for more than 40 types of HPV with a polymerase chain reaction-based system.

RESULTS

Among women with normal cytology, HPV infections peaked first in women younger than 25 years, and they peaked again at age 55 years or older with predominantly non-cancer-associated types of HPV and uncharacterized HPV types. Low-grade squamous intraepithelial lesions (LSILs) (n = 189) decreased consistently with age. The prevalence of high-grade squamous intraepithelial lesions (HSILs) (n = 128) peaked first around age 30 years and again at age 65 years or older. Seventy-three percent of LSILs were HPV positive, with HPV16 being the predominant type (16% of positive subjects). HPV was found in 89% of HSILs and 88% of cancers, with HPV16 being strongly predominant (51% and 53% of positive subjects). Virtually all HSILs and cancers had cancer-associated HPV types, with high odds ratios (ORs) and attributable fractions around 80%. Risk for HPV16 was particularly high (OR for HSILs = 320, 95% confidence interval [CI] = 97-1000; OR for cancer = 710, 95% CI = 110-4500).

CONCLUSIONS

We confirm the early decline of HPV infection with age but note increased prevalence after menopause, which could be related to a second peak of HSILs, an observation that warrants further investigation. At least 80% of HPVs involved in cervical carcinogenesis in this population have been characterized. Polyvalent vaccines including the main cancer-associated HPV types may be able to prevent most cases of cervical disease in this region.

Histological diagnosis of precancerous lesions of the stomach: a reliability study

BACKGROUND

Within the framework of a chemoprevention trial on stomach cancer, two substudies based on repeat measurement were undertaken to evaluate reliability of histological diagnoses of gastric precancerous lesions.

METHODS

A subgroup of 45 subjects received two endoscopies separated by a period of one month. The two biopsies were reviewed by a single pathologist. A second subsample of 50 subjects had a single endoscopy and the biopsy results were reviewed by two pathologists. Agreement between the two diagnoses was assessed by Cohen's Kappa and by repeat frequency.

RESULTS

When the same samples were reviewed by the pathologists involved in the trial, agreement was very high for advanced lesions (repeat frequency = 0.96 for intestinal metaplasia and 1.00 for dysplasia) but lower for less advanced lesions (repeat frequency = 0.73 for superficial gastritis and chronic gastritis, 0.65 for atrophic gastritis). When the same pathologist reviewed two sets of biopsies taken less than 2 months apart, the combination of random observer error and biopsy sampling error gave rise to quite low agreement, especially for early lesions, mainly attributable to biopsy sampling error. Comparison of diagnoses made at routine reading and at review by the same pathologist and by different pathologists showed substantial overall agreement with the exception of one pathologist for whom agreement was moderate.

CONCLUSIONS

These results confirm that misclassification of histological diagnosis may be a relevant problem in chemoprevention trials of stomach cancer, more so when baseline diagnosis is taken into account in the analysis to estimate progression and regression rates of precancerous lesions. Further, the results suggest that misclassification is limited to early lesions, while diagnostic reliability of severe lesions is quite high.

Determinants of plasma anti-oxidant vitamin levels in a population at high risk for stomach cancer

Our objective was to identify the determinants of plasma levels of anti-oxidant vitamins which have been linked with decreased risk of cancer and other chronic diseases. Correlation analyses were performed between baseline plasma levels of ascorbic acid, alpha- and beta-carotenes, cryptoxanthin, lycopene and alpha- and gamma-tocopherols and baseline information on dietary and other demographic and life-style factors among 1,364 subjects 35-69 years of age, who are participants in a chemoprevention trial on pre-cancerous lesions of the stomach in Venezuela. Males had lower levels of ascorbic acid, alpha- and beta-carotene and cryptoxanthin and higher levels of alpha-tocopherol than females. This finding was confirmed in non-smokers and non-drinkers. In females, but not in males, age was positively associated with levels of ascorbic acid, cryptoxanthin, alpha- and beta-carotene and gamma-tocopherol. Male tobacco users had lower plasma levels of ascorbic acid, alpha- and beta-carotene and cryptoxanthin than nonusers, and regular alcohol drinkers had a decreased plasma levels of beta-carotene compared with non-drinkers. Female tobacco users had lower levels of ascorbic acid and cryptoxanthin than non-users, and regular alcohol drinkers had lower levels of ascorbic acid and lycopene than non-drinkers. Frequencies of consumption of fresh fruits, fruit juice, raw vegetables and plantains showed weak positive associations with plasma levels of several vitamins studied in both sexes. Sex, age in females, tobacco and alcohol use and dietary consumption affected plasma anti-oxidant vitamin levels in this population significantly. These factors may influence the effect of anti-oxidant treatment in intervention trials.

Community health for Rwandan refugees

Health education and disease prevention programs are essential elements in every health care system. We normally envision community health programs in urban and rural settings across Canada. However, health education and disease prevention have also become a fundamental part of the health care system for refugee communities around the world.

Calibration in multi-centre cohort studies

BACKGROUND

This paper is concerned with overcoming problems caused by measurement error in multi-centre studies of diet and disease. Measurement error causes differential bias, so the information on the diet-disease relationship from different cohorts is not directly comparable. Hence this information needs to be calibrated before the data are combined in an eventual meta-analysis. We consider the design of calibration substudies. We distinguish two forms of information from a multi-centre cohort study. The first is subject-level information, which comes from the variation of disease rate within cohorts. The second is cohort-level information, which comes from the variation of disease rate between cohorts. The requirements of the calibration study are different for these two forms of information.

METHODS

Calibration is carried out by remeasuring diet in a subsample of each cohort using a standardized reference method. This reference measurement should yield unbiased estimates of habitual intake.

RESULTS

Using a criterion of efficiency, relative to a perfectly calibrated study, we show that the sample size should be a multiple of the expected number of cases in each cohort. To control for confounding, each cohort should be stratified and the ratio of sample size to number of cases should be constant within strata.

CONCLUSIONS

Since the required sample size is related not to the size of the cohort, but to the eventual number of cases of disease, calibration samples need not be prohibitively large. They should, however, be concentrated on those parts of the cohort, such as older age groups, which will yield most cases.

Seasonal variations of plasma fibrinogen and factor VII activity in the elderly: winter infections and death from cardiovascular disease

There are approximately 20,000 excess deaths from cardiovascular disease each winter in England and Wales. The reasons for the excess have not been fully elucidated. For one year, we studied 96 men and women aged 65-74 living in their own homes in order to examine seasonal variation in plasma fibrinogen and factor VII clotting activity (FVIIc), and to investigate relationships with infection and other cardiovascular-disease risk factors. Both fibrinogen and FVIIc plasma values were greater in winter with estimated winter-summer differences (confidence intervals) of 0.13 (0.05-0.20) g/L for fibrinogen and 4.2 (1.2-7.1)% of standard for FVIIc. These differences could account for 15% and 9% increases in ischaemic heart disease risk in winter respectively. After adjustment for confounding by season, fibrinogen was strongly related to neutrophil count (p < 0.0001), C-reactive protein (p < 0.0001), alpha 1-antichymotrypsin (p < 0.0001), and self-reported cough (p < 0.0001) and coryza (p = 0.0004), but not to ambient temperature. Therefore, we suggest that seasonal variation in fibrinogen might be induced by winter respiratory infections via activation of the acute phase response. Seasonal variations in the cardiovascular risk factors fibrinogen and FVIIc provide further possible explanations for the marked seasonal variation in death from ischaemic heart disease and stroke in the elderly.

Adjustment for bias due to errors in exposure assessments in multicenter cohort studies on diet and cancer: a calibration approach

An advantage of multicenter cohort studies on diet and cancer is that these may include populations over a wide range of dietary exposure. With some simplifying assumptions, the information from such multicenter studies may be divided into 1) estimated relationships within each of the separate cohorts, between individual-level measurements of dietary exposure and disease outcome, and 2) an estimated between-cohort relationship, between the mean intake measurements and mean incidence rates. Errors in the dietary exposure measurements may lead to different amounts of bias in each of these estimated relationships, in particular when dietary questionnaire methods cannot easily be standardized. A calibration approach can be used to adjust for such differences in bias. This will improve the relative weighting of within- and between-cohort components of evidence for a diet-disease association.

An outbreak of Salmonella mikawasima associated with doner kebabs

During October 1992 an increase in the number of isolates of Salmonella mikawasima, a rare serotype, was noted including a cluster of nine cases in the South West Thames region. A case control study was conducted and univariate analysis showed a statistical association between illness and eating at take-away A for cases compared with household controls (P = 0.003) and with neighbourhood controls (P = 0.0245). Cases were also more likely to have eaten kebabs than were controls or average take-away A customers, implicating doner kebabs as the most likely vehicle of infection. Plasmid profile analysis of the nine cases' isolates showed them to be indistinguishable and to be characterized by a single plasmid of approximately 60 MDa. The original source of the Salmonella mikawasima contamination was not determined, but food preparation practices for kebabs at take-away A were insufficient to protect against illness if contaminated. This outbreak was only recognized because of the unusual serotype, but could be an indication of a more widespread problem with doner kebabs.

Seasonal variation of blood pressure and its relationship to ambient temperature in an elderly population

OBJECTIVE

To measure the magnitude and timing of seasonal variation of blood pressure and related factors in the elderly living in the community, and to assess their potential impact on cardiovascular risk.

DESIGN

Prospective study; from January 1991 to February 1992 blood pressure and other variables were measured at 2-monthly intervals in each subject in their own homes.

SUBJECTS

Ninety-six men and women, age range 65-74 years, recruited from a single group general practice in Cambridge.

MAIN OUTCOME MEASURES

Seasonal variation of blood pressure, seasonal variation of prevalence of hypertension, seasonal variation of ambient temperature and body mass index.

RESULTS

Both systolic (SBP) and diastolic blood pressure (DBP) were greatest during the winter across the whole distribution of blood pressure. There was a fourfold increase in the proportion of subjects with blood pressures > 160/90 mmHg in winter compared with in summer. Regression analysis revealed highly significant seasonal differences in both SBP and DBP. After adjustment for confounding seasonal effects, a 1 degree C decrease in living-room temperature was associated with rises of 1.3 mmHg in SBP and 0.6 mmHg in DBP.

CONCLUSIONS

Seasonal variation of blood pressure is heightened in older adults and may partly explain the greater cardiovascular disease mortality of elderly subjects during the winter. The blood pressures of elderly people may be inversely related to the ambient temperature. The public health implications of these findings deserve further investigation.

Seasonal variation of serum lipids in an elderly population

We investigated seasonal variation in serum lipids in 96 volunteers aged 65-74 years who were studied at 2-monthly intervals for one year. Periodic regression analysis revealed highly significant seasonal variation in serum total cholesterol and high-density lipoprotein (HDL) cholesterol. Peak levels for both occurred in winter with corresponding summer troughs. The seasonal difference for total cholesterol was 0.32 mmol/l (95% CI 0.23-0.41, p < 0.0001) and that for HDL cholesterol 0.16 mmol/l (95% CI 0.12-0.19, p < 0.0001). Low-density lipoprotein (LDL) cholesterol was highest in winter in men only (seasonal difference 0.27 mmol/l, 95% CI 0.15-0.39, p < 0.0001), and triglycerides were significantly greater in late winter for women only (seasonal difference 0.22 mmol/l, 95% CI 0.09-0.35, p = 0.002). The timing of seasonal variation in total cholesterol and triglycerides would be consistent with a role in the seasonal variation in vascular deaths, but our finding of a relatively high HDL: total cholesterol ratio in winter makes this less likely.

Measurement error in dietary assessment: an investigation using covariance structure models. Part I

Repeated measures of diet are analysed in an attempt to discover the measurement error properties of various dietary assessment methods. It is customary in such studies to assume that one reference measurement is 'valid' (without error) but this assumption is not tenable. An alternative approach, widely used in social science, is to model the covariance matrix of the repeated measures. We apply this methodology to assessments of nitrogen intake, which is essentially equivalent to protein intake.

Measurement error in dietary assessment: an investigation using covariance structure models. Part II

In Part I we presented a covariance structure model for analysing measurement error in the assessment of nitrogen intake. In this paper we include data on urine nitrogen excretion which allows a critical assessment of the model proposed. Inclusion of urine nitrogen data produces more pessimistic estimates of the quality of dietary intake measurements and shows that previous assumptions about independence of measurement error were wrong. This underscores the need for well founded assumptions in the analysis of measurement error.

Carinal forceps biopsy via the fiberoptic bronchoscope in the routine staging of lung cancer

Main carinal biopsy was carried out in 58 consecutive patients with endobronchial (endoscopically visible) bronchogenic carcinoma. Overall, the results of the biopsy were positive in 8 of 58 patients (13.8%). The biopsy results were positive in 6 of 15 (40%) patients whose carina appeared abnormal as compared with 2 of 43 (4.7%) whose carina appeared normal (P = .0025). In those patients with subtle carinal abnormalities (carinal widening or erythema) but without gross tumor involvement, the biopsy findings were positive in 5 of 14 (36%). Unlike in previous studies, a significant percentage of positive carinal biopsy findings was associated with left upper lobe lesions. There were no complications associated with the procedure. Although the yield on blind carinal biopsies (visually normal carina) with the flexible fiberoptic bronchoscope is lower than that previously reported with the rigid bronchoscope, it remains a low-risk procedure that can spare a number of patients the morbidity and expense of more invasive surgical staging when applied as a routine part of diagnostic bronchoscopy in patients with endoscopically visible bronchogenic carcinoma.

Pervasive nature of conservative social attitudes

With a random sample of New Zealand adults, conservatives were in general not more or less inclined than liberals to like/dislike a number of food and other stimulus objects that had previously been linked to the possession of high and low self-esteem. The results provide construct validation of the Conservatism scale.