Andrew R. Willan

Determining a minimal important change in a disease-specific quality of life questionnaire

This study was carried out to determine whether the minimal important difference, in evaluative quality of life instruments which use a 7-point scale, is similar across individual domains and for both improvement and deterioration. Thirty nine adults with asthma were studied, using an 8 week cohort with assessments at 0, 4 and 8 weeks. The outcomes were the Asthma Quality of Life Questionnaire and global rating of change. For overall asthma-specific quality of life and for all individual domains (activities, emotions, symptoms), the minimal important difference of quality of life score per item was very close to 0.5 (range: 0.42-0.58); differences of approximately 1.0 represented a moderate change (range: 0.77-1.51); differences greater than 1.5 represented large changes. Changes for improvement and deterioration were very similar. The changes in quality of life score that represent a minimal important difference are very similar to those observed for other evaluative instruments. The observation that the minimal important difference is consistent across domains and for both improvement and deterioration will facilitate interpretation of results of studies examining quality of life.

Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial

BACKGROUND For 3-4% of pregnancies, the fetus will be in the breech presentation at term. For most of these women, the approach to delivery is controversial. We did a randomised trial to compare a policy of planned caesarean section with a policy of planned vaginal birth for selected breech-presentation pregnancies. METHODS At 121 centres in 26 countries, 2088 women with a singleton fetus in a frank or complete breech presentation were randomly assigned planned caesarean section or planned vaginal birth. Women having a vaginal breech delivery had an experienced clinician at the birth. Mothers and infants were followed-up to 6 weeks post partum. The primary outcomes were perinatal mortality, neonatal mortality, or serious neonatal morbidity; and maternal mortality or serious maternal morbidity. Analysis was by intention to treat. FINDINGS Data were received for 2083 women. Of the 1041 women assigned planned caesarean section, 941 (90.4%) were delivered by caesarean section. Of the 1042 women assigned planned vaginal birth, 591 (56.7%) delivered vaginally. Perinatal mortality, neonatal mortality, or serious neonatal morbidity was significantly lower for the planned caesarean section group than for the planned vaginal birth group (17 of 1039 [1.6%] vs 52 of 1039 [5.0%]; relative risk 0.33 [95% CI 0.19-0.56]; p<0.0001). There were no differences between groups in terms of maternal mortality or serious maternal morbidity (41 of 1041 [3.9%] vs 33 of 1042 [3.2%]; 1.24 [0.79-1.95]; p=0.35). INTERPRETATION Planned caesarean section is better than planned vaginal birth for the term fetus in the breech presentation; serious maternal complications are similar between the groups.

A COMPARISON OF LOW-MOLECULAR-WEIGHT HEPARIN ADMINISTERED PRIMARILY AT HOME WITH UNFRACTIONATED HEPARIN ADMINISTERED IN THE HOSPITAL FOR PROXIMAL DEEP-VEIN THROMBOSIS

BACKGROUND Patients with acute proximal deep-vein thrombosis are usually treated first in the hospital with intravenous standard (unfractionated) heparin. However, the longer plasma half-life, better bioavailability after subcutaneous administration, and more predictable anticoagulant response of low-molecular-weight heparins make them attractive for possible home use. We compared these two approaches. METHODS Patients with acute proximal deep-vein thrombosis were randomly assigned to receive either intravenous standard heparin in the hospital (253 patients) or low-molecular-weight heparin (1 mg of enoxaparin per kilogram of body weight subcutaneously twice daily) administered primarily at home (247 patients). The study design allowed outpatients taking low-molecular-weight heparin to go home immediately and hospitalized patients taking low-molecular-weight heparin to be discharged early. All the patients received warfarin starting on the second day. RESULTS Thirteen of the 247 patients receiving low-molecular-weight heparin (5.3 percent) had recurrent thromboembolism, as compared with 17 of the 253 patients receiving standard heparin (6.7 percent; P=0.57; absolute difference, 1.4 percentage points; 95 percent confidence interval, -3.0 to 5.7). Five patients receiving low-molecular-weight heparin had major bleeding, as compared with three patients receiving standard heparin. After randomization, the patients who received low-molecular-weight heparin spent a mean of 1.1 days in the hospital, as compared with 6.5 days for the standard-heparin group; 120 patients in the low-molecular-weight- heparin group did not need to be hospitalized at all. CONCLUSIONS Low-molecular-weight heparin can be used safely and effectively to treat patients with proximal deep-vein thrombosis at home.

Long-term relation between breastfeeding and development of atopy and asthma in children and young adults: a longitudinal study

BACKGROUND Breastfeeding is widely advocated to reduce risk of atopy and asthma, but the evidence for such an effect is conflicting. We aimed to assess long-term outcomes of asthma and atopy related to breastfeeding in a New Zealand birth cohort. METHODS Our cohort consisted of 1037 of 1139 children born in Dunedin, New Zealand, between April, 1972, and March, 1973, and residing in Otago province at age 3 years. Children were assessed every 2-5 years from ages 9 to 26 years with respiratory questionnaires, pulmonary function, bronchial challenge, and allergy skin tests. History of breastfeeding had been independently recorded in early childhood. FINDINGS 504 (49%) of 1037 eligible children were breastfed (4 weeks or longer) and 533 (51%) were not. More children who were breastfed were atopic at all ages from 13 to 21 years to cats (p=0.0001), house dust mites (p=0.0010), and grass pollen (p<0.0001) than those who were not. More children who were breastfed reported current asthma at each assessment between age 9 (p=0.0008) and 26 years (p=0.0008) than those who were not. Breastfeeding effects were not affected by parental history of hayfever or asthma. Multifactor analysis controlling for socioeconomic status, parental smoking, birth order, and use of sheepskin bedding in infancy, showed odds ratios of 1.94 (95% CI 1.42-2.65, p<0.0001) for any allergen positive at age 13 years, 2.40 (1.36-4.26, p=0.0003) for current asthma at 9 years, and 1.83 (1.35-2.47, p<0.0001) for current asthma at 9-26 years by repeated-measures analysis. INTERPRETATION Breastfeeding does not protect children against atopy and asthma and may even increase the risk.

Interventions to reduce unintended pregnancies among adolescents: systematic review of randomised controlled trials

Abstract Objective: To review the effectiveness of primary prevention strategies aimed at delaying sexual intercourse, improving use of birth control, and reducing incidence of unintended pregnancy in adolescents. Data sources: 12 electronic bibliographic databases, 10 key journals, citations of relevant articles, and contact with authors. Study selection: 26 trials described in 22 published and unpublished reports that randomised adolescents to an intervention or a control group (alternate intervention or nothing). Data extraction: Two independent reviewers assessed methodological quality and abstracted data. Data synthesis: The interventions did not delay initiation of sexual intercourse in young women (pooled odds ratio 1.12; 95% confidence interval 0.96 to 1.30) or young men (0.99; 0.84 to 1.16); did not improve use of birth control by young women at every intercourse (0.95; 0.69 to 1.30) or at last intercourse (1.05; 0.50 to 2.19) or by young men at every intercourse (0.90; 0.70 to 1.16) or at last intercourse (1.25; 0.99 to 1.59); and did not reduce pregnancy rates in young women (1.04; 0.78 to 1.40). Four abstinence programmes and one school based sex education programme were associated with an increase in number of pregnancies among partners of young male participants (1.54; 1.03 to 2.29). There were significantly fewer pregnancies in young women who received a multifaceted programme (0.41; 0.20 to 0.83), though baseline differences in this study favoured the intervention. Conclusions: Primary prevention strategies evaluated to date do not delay the initiation of sexual intercourse, improve use of birth control among young men and women, or reduce the number of pregnancies in young women.

In Search of Power and Significance: Issues in the Design and Analysis of Stochastic Cost-effectiveness Studies in Health Care

Application of techniques such as cost-effectiveness analysis (CEA) is growing rapidly in health care. There are two general approaches to analysis: deterministic models based upon assumptions and secondary analysis of retrospective data, and prospective stochastic analyses in which the design of a clinical experiment such as randomised controlled trial is adapted to collect patient-specific data on costs and effects. An important methodological difference between these two approaches is in the quantification and analysis of uncertainty. Whereas the traditional CEA model utilizes sensitivity analysis, the mean-variance data on costs and effects from a prospective trial presents the opportunity to analyze cost-effectiveness using conventional inferential statistical methods. In this study we explored some of the implications of moving economic appraisal away from deterministic models and toward the experimental paradigm. Our specific focus was on the feasibility and desirability of constructing statistical tests of economic hypotheses and estimation of cost-effectiveness ratios with associated 95% confidence intervals. We show how relevant variances can be estimated for this task and discuss the implications for the design and analysis of prospective economic studies.

Laboratory diagnosis of iron-deficiency anemia: an overview.

BACKGROUND AND METHODS To determine the diagnostic values of laboratory tests used in the diagnosis of iron-deficiency anemia, the authors conducted a systematic overview of the relevant literature. Computerized searches of the MEDLINE database yielded 1,179 potentially relevant citations. Fifty-five studies included the results of laboratory tests and histologic examination of the bone marrow for at least 50% of an identifiable patient group. In these 55 studies, quality was assessed and descriptive information concerning the study populations, the tests conducted, and the results was extracted, all in duplicate. RESULTS Serum ferritin radioimmunoassay was by far the most powerful test, with an area under the receiver operating characteristic curve of 0.95. Test properties differed for populations of patients with inflammatory, liver, or neoplastic disease and patients without these conditions. Likelihood ratio lines, which allow precise interpretation of results across the entire range of ferritin concentration values, were constructed for the individual populations. CONCLUSION Serum ferritin radioimmunoassay is an extremely powerful test for the diagnosis of iron-deficiency anemia and, appropriately interpreted, can be applied to the complete range of patients.

The prognosis for live birth among untreated infertile couples

OBJECTIVE To estimate the likelihood of live birth and the impact of prognostic factors among untreated infertile couples. DESIGN Cohort follow-up study. SETTING Eleven infertility clinics in academic medical centers. PATIENTS Two thousand one hundred ninety-eight couples with infertility of > 1 year in duration. Untreated months of observations on 873 untreated couples (18,364 months) were combined with observations before the first treatment among 1,325 treated couples (9,761 months). INTERVENTIONS Diagnostic tests as indicated. MAIN OUTCOME MEASURES Time to conception for live birth; effect of baseline characteristics estimated by means of proportional hazards analysis. RESULTS There were 263 live birth conceptions during 28,125 months of untreated observations. The cumulative rate of conceptions leading to live birth was 14.3% at 12 months. The relevant prognostic factors were pregnancy history, duration of infertility, female partners age, male defect, endometriosis, and tubal disease. A prediction score based on these factors would be accurate in approximately 62% of cases. CONCLUSIONS The estimation of live birth among untreated infertile couples is sufficiently accurate to be useful in the clinical management of infertility and in the planning of clinical trials.

Laboratory diagnosis of iron-deficiency anemia

Background and methods:To determine the diagnostic values of laboratory tests used in the diagnosis of iron-deficiency anemia, the authors conducted a systematic over-view of the relevant literature. Computerized searches of the MEDLINE database yielded 1,179 potentially relevant citations. Fifty-five studies included the results of laboratory tests and histologic examination of the bone marrow for at least 50% of an identifiable patient group. In these 55 studies, quality was assessed and descriptive information concerning the study populations, the tests conducted, and the results was extracted, all in duplicate.Results:Serum ferritin radioimmunoassay was by far the most powerful test, with an area under the receiver operating characteristic curve of 0.95. Test properties differed for populations of patients with inflammatory, liver, or neoplastic disease and patients without these conditions. Likelihood ratio lines, which allow precise interpretation of results across the entire range of ferritin concentration values, were constructed for the individual populations.Conclusion:Serum ferritin radioimmunoassay is an extremely powerful test for the diagnosis of iron-deficiency anemia and, appropriately interpreted, can be applied to the complete range of patients.

Confidence intervals for cost-effectiveness ratios: an application of Fieller's theorem.

Application of cost-effectiveness analysis (CEA) is growing rapidly in health care. Two general approaches to analysis are differentiated by the type of data available: (i) deterministic models based upon secondary analysis of retrospective data from one or more trials and other sources; and (ii) stochastic analyses in which the design of a randomized controlled trial is adapted to collect prospectively patient-specific data on costs and effectiveness. An important methodological difference between these two approaches is in how uncertainty is handled. Deterministic CEA models typically rely upon sensitivity analysis to determine the robustness of findings to alternative assumptions, whereas stochastic (CEA) analysis, as part of prospective studies, permits the use of conventional statistical methods on the cost and effectiveness data for both inference (hypothesis testing) and estimation. This paper presents a procedure for the statistical analysis of cost-effectiveness data, with specific application to those studies for which effectiveness is measured as a binary outcome. Specifically, Fiellers Theorem was used to calculate confidence intervals for ratios of the two random variables of between-treatment differences in observed costs and effectiveness, i.e. the incremental cost-effectiveness ratio.