Statistics in respiratory medicine. 2. Repeatability and method comparison.
Explore the source record for details and available documents.
Methods↗
Biomedical subjects
S Chinn
Publications and source records attributed to S Chinn.
Statistics in respiratory medicine. 3. Scale, parametric methods, and transformations.
Explore the source record for details and available documents.
Analysis of Variance↗
Reinforcement of incontinent stools in the treatment of encopresis.
In this paper, fecal incontinence and constipation were treated in two mentally retarded individuals using a novel intervention wherein incontinent stools were first rewarded in order to increase the frequency of bowel movements. This intervention was implemented only after more traditional pediatric and behavioral treatments were unsuccessful. The reinforcement of incontinent stools procedure resulted in an increase in both the frequency and the number of continent stools for both individuals. Hypotheses regarding the failure of previous treatments and the success of the current treatment are discussed as well as directions for future research.
Adolescent↗
Has the prevalence of asthma increased in children? Evidence from the national study of health and growth 1973-86.
OBJECTIVES: To estimate changes in the prevalence of reported symptoms of respiratory disease and reported diagnoses of asthma and bronchitis in primary school children in England between 1973 and 1986. DESIGN: Mixed longitudinal survey. SETTING: Representative sample of English primary schools in 22 areas. PARTICIPANTS: 15,000 Boys and 14,156 girls each studied at least once between 1973 and 1986. DATA COLLECTED: Whether, according to the parent or guardian, the child had wheezed, wheezed on most days or nights, or had attacks of bronchitis or asthma in the past year. RESULTS: Within age groups trends in successive annual cohorts showed an increasing prevalence of asthma for each annual birth cohort (boys, 6.9%, p less than 0.001; girls, 12.8%, p less than 0.001) and of wheeze on most days or nights (boys, 4.3% per cohort, p less than 0.001; girls, 6.1% per cohort, p less than 0.001) and a falling prevalence of bronchitis (boys, -4.7% per cohort, p less than 0.001; girls, -5.8% per cohort, p less than 0.001). There was a smaller increase in the prevalence of wheeze whether or not it occurred on most days or nights, and this increase was significant only among the girls (boys, 1.0% per cohort, p greater than 0.05; girls, 1.7% per cohort, p less than 0.05). Although the rate of increase of "asthma" was greater than the rate of decrease in "bronchitis," the baseline prevalence of asthma was much lower than that of bronchitis, and the total proportion of children with either diagnosis declined slightly over the whole period. The main change was an increase in the proportion of children whose parents stated that they had persistent wheeze and yet did not have a report of either "asthma" or "bronchitis." CONCLUSIONS: These results suggest that there has been a true increase in morbidity that is not simply due to changes in diagnostic fashion. The increase is large enough to explain much if not all of the increase in admission to hospital and mortality, and it underlines the importance of an understanding of the aetiology of asthma in tackling the causes of the recent increase.
Asthma↗
The assessment of methods of measurement.
Criteria are given for the choice of scale prior to estimation of repeatability. Recommendations of Bland and Altman should then be used for expressing repeatability and agreement of methods of measurement on the same scale. Repeatability of measurements on different scales should be compared using the appropriate ratio of variances, or intraclass correlation coefficient. A reference range for diagnosis requires a high ratio of between-subject variation to total variation. The index of separation between diseased and healthy subjects should be used whenever possible. Changes within patients should be compared with reference change ranges, and not against the diagnostic range.
Diagnosis↗
Associations of excessive irritability with common illnesses and food intolerance.
In a national study of almost 7000 primary school children, parents' perceptions were used to test the hypothesis that the child's irritability was associated with food intolerance independently of other symptoms. After adjustment in a multiple regression analysis for asthma or wheeze, cough, eczema, hives, diarrhoea and vomiting, rhinitis, hay fever and headache, and the social factors of father's social class, maternal education and maternal age, a highly significant association between perception of food intolerance and irritability (P less than 0.001) remained. Though we cannot rule out that irritable children's parents could be biased towards diagnosing food intolerance the possibility that some children do have behavioural disturbance associated with reactions to food needs to be explored further, preferably with a double blind challenge assessment.
Affect↗
Sleep habits and height at ages 5 to 11.
Shorter durations of slow wave sleep and lower growth hormone responses have been reported in children with short stature caused by psychosocial deprivation. We investigated whether lower total sleep duration was associated with shorter stature in a sample of children taking part in the National Study of Health and Growth. Parental responses to a self administered questionnaire were used to estimate usual times for going to sleep at night and usual times for waking in the morning for 5145 children aged 5 to 11 years of age. After adjusting for the effects of other variables known to be associated with height, it was shown that there was a weak negative association between sleep duration and height. It is concluded that variation in sleep duration between children is unlikely to have an important influence on growth.
Body Height↗
BCG in management of transitional cell carcinoma invasive to prostate.
Transitional cell carcinoma invading the prostate is a difficult problem to manage. Transurethral resection of the bladder neck followed by intravesical bacillus Calmette-Guerin (BCG) is one suggested option.
Aged↗
The repeatability and validity of respiratory resistance measured by the forced oscillation technique.
Measurement of forced expiratory volume in one second (FEV1) is commonly used in bronchial challenge testing in laboratory and epidemiological studies but has certain disadvantages. We have therefore studied the repeatability and validity of a derived measurement of respiratory resistance (Ros) obtained with the Siregnost FD5 impedance oscillometer (Siemens). Repeatability was estimated in 25 non-asthmatics and 28 asthmatics and compared with that of PEFR, FEV1 and specific conductance. PEFR and FEV1 were the most repeatable. Repeatability for measurements with the oscillometer as measured by the intraclass correlation coefficient for a single measurement was 0.75 for non-asthmatics, but was less good for asthmatics (intraclass correlation coefficient 0.56). To assess the validity of respiratory resistance measured during a histamine bronchial challenge test a sample of 20 hospital personnel was studied on four occasions, FEV1 and Ros both being measured twice. Ros changed at lower doses of histamine than FEV1, but the intraclass correlation coefficient for repeatability of change only reached 0.6 at an absolute dose of 3.41 mumol histamine. The estimated provocation dose of histamine producing a 35% fall in Ros was 8.70 mumol (95% range for a single measurement +/- 1.11 doubling doses), that producing a 10% fall in FEV1 was 8.32 mumol (95% range +/- 1.04 doubling doses) and a 20% fall in FEV1 11.48 mumol (95% range +/- 1.11 doubling doses). Measurements obtained with the Siregnost FD5 oscillometer are repeatable. The use of Ros during bronchial challenge testing is valid, but shows insufficient advantage over FEV1 to support its use in epidemiological studies.
Adult↗
What symptoms predict the bronchial response to histamine? Evaluation in a community survey of the bronchial symptoms questionnaire (1984) of the International Union Against Tuberculosis and Lung Disease.
A questionnaire developed by the International Union against Tuberculosis and Lung Disease (IUATLD) to assess bronchial symptoms has been tested for its ability to predict the bronchial response to histamine in adults aged 18-64 years living in two areas of southern England. A number of questions were found to be independently associated with increased reactivity in the first randomly selected half of the subjects. These symptoms included wheeze, waking at night with shortness of breath, tightness in the chest or shortness of breath when exposed to animals, dust or feathers and the non-specific symptom of persistent problems with breathing. A predictive score based on these symptoms was more sensitive and only slightly less specific than the question on wheeze alone in predicting the response to histamine in the second half of the subjects. Questions about asthma though more specific were considerably less sensitive than either. Symptoms did not differentiate between reactivity associated with positive skin tests and that associated with smoking.
Adult↗
A qualitative and quantitative study of the cellular fatty acids of 'Streptococcus milleri' with capillary gas chromatography.
Fatty acid analysis was done with GC and GC-MS on 21 strains of 'Streptococcus milleri', representative of the various proposed species. Although no qualitative differences were found in the fatty acid profiles, discriminant analysis of the quantitative data revealed three groups. Streptococcus anginosus and Streptococcus constellatus were indistinguishable but separated from the other two groups which comprised Streptococcus intermedius, with a wide fermentation pattern and Streptococcus intermedius with a narrow fermentation pattern. Three of the strains could be distinguished from the others by a 'fingerprint' of a particularly prominent fatty acid peak. The results support the suggestion that there is more than one species in this group of organisms and that the technique might be of value in epidemiological investigations of 'S. milleri'.
Fatty Acids↗
Need for new reference curves for height.
Data from the National Study for Health and Growth, on children aged from 4.0 to 12.0 years measured in 1972, 1985, and 1986 were used to assess whether new growth standards are required, and which subgroups of children might require separate standards. The change over this period, from just over half a centimetre in the youngest girls to over a centimetre in the oldest boys, warrants the use of revised reference curves, which are also needed for Afro-Caribbean children.
Body Height↗
School meals, school milk and height of primary school children in England and Scotland in the eighties.
In a nutritional surveillance system of primary school children in England and Scotland we assessed the possible effects on height gain of changes in school meals and school milk policies following the 1980 Education Act (No. 2). Mean height and height gain were estimated separately for English and Scottish samples from 1982 to 1984, and for a selective sample of inner city areas with a high proportion of ethnic minorities from 1983 to 1985 in children from 5.00 to 9.99 years. Children receiving free school meals were smaller than children paying for school meals or receiving a meal prepared elsewhere, while children receiving free school milk were of similar stature to other children in the study. The rate of growth was assessed in children receiving school meals or lunches prepared at home, and in those for whom arrangements changed during the study period; it was also assessed in those children for whom school milk was available, not available, or for whom the provision changed. No consistent association was found between provision of school meals or school milk and the rate of growth in the three samples studied when stratified according to poverty status and ethnic background. We conclude that this observational study does not provide any evidence that the current availability of school meals or school milk increases the rate of growth of primary school children in any social stratum.
Animals↗
Relation between all cause standardised mortality ratios and two indices of deprivation at regional and district level in England.
The use of mortality data in the form of standardised mortality ratios (SMRs) to measure the need for health care resources in the Resource Allocation Working Party (RAWP) formula in England has been criticised for underestimating the wider effects of adverse socioeconomic conditions on need, particularly in inner city areas. To assess this criticism, we explored the relationships at NHS Regional and District levels in England between two indicators of illness from the 1981 Census, two contrasting indices of deprivation based on the 1981 Census (the Jarman 8 Underprivileged Area (UPA) score and Townsend's Index of Material Deprivation) and their constituent variables, and all cause SMRs for 1982-3. All cause SMRs were highly correlated at Regional and District level with permanent and temporary sickness rates. At Regional level, three of the Thames Regions showed relatively high deprivation scores in relation to their SMRs, in comparison to the remaining Regions where the relative level of deprivation closely matched the Region's mortality ranking. District level analyses of the relations between SMRs and the deprivation indices and their constituent variables showed that the Thames/non-Thames dichotomy was accounted for by the 14 Districts in inner London. These findings suggest that although there may be a prima facie case for including an allowance for deprivation in RAWP, it is still not clear how the deprivation variables available in the Census relate empirically to the need for additional health service resources. The analysis raises questions about the appropriate definition of need in this context and whether the Census is a suitable source for the construction of a deprivation weighting for use in national RAWP.
Community Health Services↗
Effect of changing dietary sodium on the airway response to histamine.
The airway response to histamine has been shown to be related to the 24 hour urinary excretion of sodium. To assess whether this relation is likely to represent a direct causal association a randomised double blind crossover trial of slow sodium (80 mmol/day) was compared with placebo in 36 subjects having a low sodium diet. The dose of histamine causing a 20% fall in FEV1 (PD20) was 1.51 doubling doses lower when the men were taking sodium than when they were taking placebo (p less than 0.05). On the basis of PD10 values, the difference in men was 1.66 doubling doses of histamine (p less than 0.05). There was no corresponding effect in women. Regressing PD10 against urinary excretion of electrolytes with data from the two occasions during the trial and the measurements made before the trial showed a significant association with sodium excretion after allowance had been made for any effect associated with potassium or creatinine excretion, the latter being a marker of the completeness of the urine collection. Again there was no corresponding effect among women. These findings are compatible with the differences in regional mortality data for England and Wales, which show a relation between asthma mortality and regional per person purchases of table salt for men but not for women.
Adult↗
The distribution of peak expiratory flow variability in a population sample.
Although serial peak expiratory flow (PEF) measurements are often used to assess the variability of airflow obstruction, the range of values to be expected in the general population has never been defined, nor is there any consensus as to how PEF variability should be expressed. We have compared PEF recordings made by 121 subjects selected at random from the population of a small town (Group A) and 221 subjects selected because of wheeze in the last year (Group B). Subjects were asked to record PEF every 2 h during waking hours for 7 days using a mini-Wright peak flow meter. Seven indices of PEF variability were derived for each subject and the range for each index determined. All indices showed a positively skewed distribution in the random sample. Two variability indices, standard deviation percent mean and amplitude percent mean, provided the greatest separation between subjects with both a diagnosis of asthma and wheeze in the last year and subjects with neither feature and also provided the highest intra-class correlation coefficients. We conclude that amplitude percent mean and standard deviation percent mean provide the best means of expressing PEF variability for epidemiological purposes, but that amplitude percent mean is more easily derived and appears to be the index of choice. PEF variability has a continuous distribution in the general population and no clear-cut division between asthmatic and nonasthmatic subjects can be defined.
Adolescent↗
Validity and repeatability of the IUATLD (1984) Bronchial Symptoms Questionnaire: an international comparison.
The International Union against Tuberculosis and Lung Disease (IUATLD) Bronchial Symptoms Questionnaire (1984) was developed for use in studies of asthma and its reliability measured in an earlier survey in England. The association of the symptoms elicited by this questionnaire to bronchial response to histamine has also been described. This paper presents the results of studies of the questionnaire in four clinical centres in Europe. The reliability of the questionnaire and its ability to predict the bronchial response to histamine were compared for English, Finnish, French and German translations of the questionnaire in samples of diagnosed asthmatics and controls in Nottingham, Berlin, Helsinki and Paris. The answers to questions showed good repeatability, especially in Finland and Germany, particularly those questions on asthma and wheeze. The most sensitive symptom for predicting hyperresponsiveness was the question on wheeze, the most specific questions were those on waking at night with shortness of breath (Paris and Nottingham) and morning tightness (Helsinki and Berlin). This study shows that the IUATLD (1984) questionnaire may provide useful, valid and comparable data even in translation but these studies will need to be repeated in representative samples before such a possibility is accepted as fully demonstrated.
Adult↗
Longitudinal studies: objectives and ethical considerations.
Studies discussed in this issue have in common that subjects are followed-up, the majority of studies with data collection on at least two occasions. However the need for follow-up in terms of the study hypotheses should be assessed carefully, because such studies are costly and more difficult to carry-out and analyse than cross-sectional studies. Studies are classified according to the nature of the explanatory and outcome variables, whether fixed for each subject, a transitional age, or a changing measurement or characteristic. The information that can be obtained is discussed for each combination of explanatory and outcome variables. Time trends are also considered. The different requirements of reference ranges are included, and ethical considerations particular to follow-up studies are briefly described.
Age Factors↗