The Cutaneous Assessment Tool: development and reliability in juvenile idiopathic inflammatory myopathy.

OBJECTIVES: Clinical care and therapeutic trials in idiopathic inflammatory myopathies (IIM) require accurate and consistent assessment of cutaneous involvement. The Cutaneous Assessment Tool (CAT) was designed to measure skin activity and damage in IIM. We describe the development and inter-rater reliability of the CAT, and the frequency of lesions endorsed in a large population of juvenile IIM patients. METHODS: The CAT includes 10 activity, 4 damage and 7 combined lesions. Thirty-two photographic slides depicting IIM skin lesions were assessed by 11 raters. One hundred and twenty-three children were assessed by 11 paediatric rheumatologists at 10 centres. Inter-rater reliability was assessed using simple agreements and intra-class correlation coefficients (ICC). RESULTS: Simple agreements in recognizing lesions as present or absent were generally high (0.5-1.0). ICCs for CAT lesions were moderate (0.4-0.75) in both slides and real patients. ICCs for the CAT activity and damage scores were 0.71 and 0.81, respectively. CAT activity scores ranged from 0 to 44 (median 7, potential range 0-96) and CAT damage scores ranged from 0 to 13 (median 1, potential range 0-22). The most common cutaneous lesions endorsed were periungual capillary loop changes (63%), Gottron's papules/sign (53%), heliotrope rash (49%) and malar/facial erythema (49%). CONCLUSIONS: Total CAT activity and damage scores have moderate to good reliability. Assessors generally agree on the presence of a variety of cutaneous lesions. The CAT is a promising, semi-quantitative tool to comprehensively assess skin disease activity and damage in IIM.

Lot consistency as an equivalence problem.

One requirement for licensure of a vaccine in the United States is demonstration by the manufacturer of consistently produced lots of vaccine. Demonstration of consistency of manufacturing can be viewed as a multigroup equivalence problem. The standard statistical procedures for evaluating equivalence assume normally distributed data and define equivalence margins with respect to group means. As an alternative approach, we define a measure of the similarity among group distributions and their nonparametric estimators. Through computer simulations we explore the statistical properties of an equivalence test based on this estimator and compare them to the standard methods. Preliminary work suggests that a test of similarity can be useful in demonstrating equivalence when distributions are not normal or when there are differences in scale or shape. It appears to detect departures from equivalence that are not reflected by differences among group means.

Repeated measures with zeros.

Consider repeated measures data with many zeros. For the case with one grouping factor and one repeated measure, we examine several models, assuming that the nonzero data are roughly lognormal. One of the simplest approaches is to model the zeros as left-censored observations from the lognormal distribution. A random effect is assumed for subjects. The censored model makes a strong assumption about the relationship between the zeros and the nonzero values. To check on this, you can instead assume that some of the zeros are 'true' zeros and model them as Bernoulli. Then the other values are modeled with a censored lognormal. A logistic model is used for the Bernoulli p, the probability of a true nonzero. The fit of the pure left-censored lognormal can be assessed by testing the hypothesis that p is 1, as described by Moulton and Halsey. The model can also be simplified by omitting the censoring, leaving a logistic model for the zeros and a lognormal model for the nonzero values. This is approximately equivalent to modeling the zero and nonzero values separately, a two-part model. In contrast to the censored model, this model assumes only a slight relationship (a covariance component) between the occurrence of zeros and the size of the nonzero values. The models are compared in terms of an example with data from children's private speech.

Analysis of studies to evaluate immune response to combination vaccines.

The development and evaluation of new combination vaccines is an important public health endeavor. Trials to evaluate these vaccines are customarily designed and analyzed as noninferiority studies. We explain the concept of noninferiority and highlight important issues that can be challenging in the statistical evaluation of these vaccines. Topics covered include end points, hypotheses, and analyses for comparing geometric mean concentrations (or titers) of antibody and proportion of vaccine recipients responding; covariate adjustment; the problem of multiplicity and its impact on sample size; and choice of a meaningful difference to rule out.

Validation of the Childhood Health Assessment Questionnaire in the juvenile idiopathic myopathies. Juvenile Dermatomyositis Disease Activity Collaborative Study Group.

OBJECTIVE: To examine the validity of the Childhood Health Assessment Questionnaire (CHAQ) in patients with juvenile idiopathic inflammatory myopathy (IIM). METHODS: One hundred fifteen patients were enrolled in a multicenter collaborative study, during which subjects were assessed twice, 7-9 months apart. Physical function was measured using the CHAQ. Internal reliability was assessed using adjusted item-total correlations and item endorsement rates. Construct validity was assessed by comparing predicted and actual correlations of the CHAQ with other measures of physical function and disease activity. Responsiveness was assessed by calculating effect size (ES) and standardized response mean (SRM) in a group of a priori defined "improvers." RESULTS: Item-total correlations were high (rs range = 0.35-0.81), suggesting all items were related to overall physical function. Manual muscle testing and the Childhood Myositis Assessment Scale correlated moderate to strongly with the CHAQ (r = -0.64 and -0.75, both p < 0.001). Moderate correlations were also seen with the physician global assessment of disease activity (rs = 0.58, p < 0.001), parent global assessment of overall health (rs = -0.65, p < 0.001), Steinbrocker function class (rs = 0.69, p < 0.001), and global skin activity (rs = 0.40, p < 0.001), while global disease damage and skin damage had low correlations (rs = 0.13 and 0.07, p > or =0.17). Responsiveness of the CHAQ was high, with ES = 1.05 and SRM = 1.20. CONCLUSION: In this large cohort of patients with juvenile IIM, the CHAQ exhibited internal reliability, construct validity, and strong responsiveness. We conclude that the CHAQ is a valid measure of physical function in juvenile IIM, appropriate for use in therapeutic trials, and potentially in the clinical care of these patients.

Comparisons of two-part models with competitors.

Two-part models arise when there is a clump of 0 observations in a distribution of continuous non-negative responses. Several methods for comparing two such distributions are available. These include the straightforward application of the z-test (or t-test), the Wilcoxon-Mann-Whitney rank sum test, the Kolmogorov-Smirnov test, and three tests that use a 2 degree of freedom chi(2) test based on the sum of the test for equality of proportions and a conditional chi(2) test for the continuous responses. This conditional test may be the z-test, the rank sum test, or the chi(2) corresponding to the Kolmogorov-Smirnov test. This study compares the size and power of several of these methods. All tests have the appropriate distribution under the null hypothesis if the distribution of the continuous part has finite moments. If it does not, the z-test has no power to detect any alternatives. It is found that the 2 d.f. tests are superior to the others when the larger proportion of 0 values corresponds to the population with the larger mean. If the reverse holds, the difference in the proportion of zeros reinforces the difference in means and some single-part models (the rank sum or Kolmogorov-Smirnov) do best. In those cases, the two-part models are not far behind, although statistically significantly poorer with respect to power. Published in 2001 by John Wiley & Sons, Ltd.

Power and sample size requirements for two-part models.

Two-part models assume the data has a probability mass at zero and a continuous response for values greater than 0. We construct tests based on the proportion of zeros and the difference among the positive values of a response giving rise to a two degree of freedom chi(2) test. This note gives the non-centrality parameter and finds the power for the test. The power is compared to the simulation results in a recent paper. We derive sample size calculations. We note that some modifications of this procedure may provide better tests if the non-zero responses are discrete. Published in 2001 by John Wiley & Sons, Ltd.

Intent-to-treat analysis and preventive vaccine efficacy.

While the intent-to-treat (ITT) concept has often been discussed and debated in the literature with respect to randomized, placebo-controlled therapeutic trials, there has been little discussion of this issue in the context of preventive vaccine efficacy trials. ITT analysis has traditionally played a minor role (if any) in the latter trials. This paper discusses the ITT approach to analysis in randomized superiority trials of preventive vaccine efficacy, using clinical endpoints. Data are presented from published literature as well as from a simple mathematical model. The data suggest that when compliance and efficacy are high, both ITT and "per-protocol" approaches generally lead to similar conclusions regarding the acceptability of a vaccine for use in a population. However, when compliance is low, the ITT and per-protocol estimates of vaccine efficacy can be widely disparate. ITT and per-protocol analyses address unique and relevant scientific questions, and often both will be informative in evaluating preventive vaccines.

Assessing screening tests: extensions of McNemar's test.

We address the problem of comparing a new screening test to a currently available screening test in the absence of a gold standard. When both tests are given to each participant in a clinical trial, the usual analytical approach is to apply McNemar's test for equality of the off-diagonal probabilities, with rejection of the null hypothesis implying that the tests differ. For assessing equivalence, however, we consider a compound null hypothesis that the new test gives either fewer or more positive results than the standard. If both parts of this hypothesis are rejected, we assert equivalence in the rate of positive responses. We propose an extension of McNemar's test for this situation. A companion step is to construct a confidence interval for the ratio of the marginal probabilities and assert equivalence if the interval is sufficiently small. It is also important that the tests agree a large proportion of the time. This can be verified with a complementary two-tailed binomial test. Another situation arises when there is a gold standard for disease diagnosis, and we wish to compare the sensitivity and specificity of two screening tests. We show that a 2 degrees-of-freedom chi-square test based on two McNemar-like tables is an appropriate test.

The modified Rodnan skin score is an accurate reflection of skin biopsy thickness in systemic sclerosis.

OBJECTIVE: To test the ability of the modified Rodnan skin score to reflect skin thickness in skin biopsies from 141 patients with systemic sclerosis (SSc) obtained at entry during a prospective, double blind study of ketanserin versus placebo in SSc. METHODS: Punch skin biopsies (4 mm) were obtained from the dorsal surface of the distal forearm of 141 patients. Biopsy specimens were trimmed and weighed (wet weight) and then desiccated and reweighed (dry weight). Skin score was recorded for 17 areas, graded 0-4+, while edema was graded 0-4+ in 10 of the same sites using finger pressure. RESULTS: Total skin score correlated with wet weight (r = 0.553) and dry weight (r = 0.517) of the skin biopsies. Local skin score from the biopsied forearm also correlated with wet and dry weight (r = 0.536 and 0.530, respectively). Dry weight as a percentage of wet weight was the same for diffuse cutaneous SSc (dSSc) and limited cutaneous SSc (lSSc) (30.7% for both, NS), despite increased wet weight in patients with dSSc versus lSSc (17.75 vs 13.03 g; p < 0.001). Edema scores correlated poorly both with wet weight (r = 0.069) and dry weight (r = 0.169). CONCLUSION: Total and forearm skin score correlates well with both wet and dry forearm skin biopsy weight from forearm biopsies, indicating that skin score reflects the underlying pathology of SSc. Further, the percentage of dry to wet weight is similar for lSSc and dSSc, supporting the usefulness of skin score in differentiating SSc disease subtypes.

Sensitivity, specificity, and vaccine efficacy.

We study the effects of imperfect sensitivity and specificity on estimation of vaccine efficacy (VE). We show that a specificity of <1.0 (i.e., some noncases are classified as cases) can reduce the value of the estimate of VE. One may attempt to adjust the estimate by accounting for the values of sensitivity and specificity. This increases the variance of the estimate of VE and leads to larger sample size requirements. Alternatively, one may adopt the diagnosis method as the disease definition.

Development of validated disease activity and damage indices for the juvenile idiopathic inflammatory myopathies: I. Physician, parent, and patient global assessments. Juvenile Dermatomyositis Disease Activity Collaborative Study Group.

OBJECTIVE: To determine the reliability, content validity, and responsiveness of physician global assessments of disease activity and damage in the juvenile idiopathic inflammatory myopathies (IIM), and to investigate concordance among physician, parent, and patient global ratings. METHODS: Sixteen pediatric rheumatologists rated 10 juvenile IIM paper patient cases for global disease activity and damage, and assessed the importance of 51 clinical and laboratory parameters in formulating their global assessments. Then, 117 juvenile IIM patients were enrolled in a protocol to examine the relationship between Likert and visual analog scale global assessments, their sensitivity to change, and the comparability of physician, parent, and patient global ratings. RESULTS: Pediatric rheumatologists demonstrated excellent interrater reliability in their global assessments of juvenile IIM disease activity and damage (97.7% and 94.7% agreement among raters, respectively), and agreed on a core set of clinical parameters in formulating their judgments. Likert scale ratings correlated with those on a visual analog scale, and both were comparable in responsiveness (standardized response means -0.56 for disease activity, 0.02 [Likert] and 0.14 [visual analog] for damage, measured over 8 months). Parent global ratings of disease activity correlated with physician assessments, but were not colinear (Spearman's correlation [r] = 0.41-0.45). Patient global disease activity assessments correlated with those done by parents (r = 0.57-0.84) and physicians (r = 0.37-0.63), but demonstrated less responsiveness (standardized response means -0.21 and -0.12, respectively, over 8 months). CONCLUSION: Physician global assessments of juvenile IIM disease activity and damage demonstrated high interrater reliability and were shown to be comprehensive measures. Both physician and parent disease activity assessments should be considered valuable as quantitative measures for evaluating therapeutic responses in juvenile IIM patients.

Discriminant diagnostics.

I discuss diagnostic methods for discriminant analysis. The equivalence with linear regression is noted and regression diagnostics are considered. The leverage is a function of the linear discriminant function and the Mahalanobis distance of the observation from the group mean. The distribution of this distance is approximately chi-square with degrees of freedom equal to the number of variables. Standard tests may be used to examine the normality assumption. Some examples are given.

Tutorial in biostatistics. Designing studies for dose response.

'Dose response' refers to the regression of a response on a stimulus. We review a number of options for dose-response designs, and compare various designs which may be used in practice. We start with two group designs. Next, we introduce basic optimal approximate design theory for simple linear and quadratic regression illustrating different criteria of optimality and their effect on the allocation of the levels of the dose. Then we obtain the efficiencies of these optimal approximate designs and some simple designs which have intuitive appeal (symmetry, equal spacing of treatments, reduced numbers of observations at the highest and lowest doses).

Comparative short-term health responses to sulfur dioxide exposure and other common stresses in a panel of asthmatics.

We studied 14 unmedicated sulfur dioxide (SO2)-sensitive asthmatics to test the hypothesis that SO2 exacerbates asthma more than other everyday respiratory stressors. In Phase I, subjects underwent controlled exposures to 0.0, 0.5, and 1.0 ppm SO2 with light, medium, and heavy exercise (average ventilation 30, 36, and 43 l/min, respectively). Lung function, symptoms of asthma, and psychophysical (stamina) changes were measured. Function, symptom, and stamina responses correlated modestly. Increasing SO2 had stronger unfavorable effects than increasing exercise. In Phase II, subjects performed eight different physical tasks in SO2-free ambient air while symptoms and stamina were measured. Fast stair-climbing evoked symptoms similar to the effects of 0.5 ppm SO2/light exercise, while stamina reduction was comparable to 0.5 ppm SO2/heavy exercise. In Phase III, subjects recorded time-activity patterns, symptoms, and stamina during randomly selected intervals on a typical weekday and weekend day. Most reported activities were sedentary. Infrequent, strenuous Phase III exercise increased symptoms more than did 0.5 ppm SO2/light exercise, but with less effect on stamina. We conclude that for typical mild asthmatics, ten-minute SO2 exposures at concentrations > 0.5 ppm and ventilation > 30 l/min can cause short-term asthma manifestations more intense than those usually experienced from everyday stresses without SO2 exposure.

Concordant and discrepant views of patients' physical functioning.

OBJECTIVE: Although considerable research has been done on patient-physician interaction, few studies have examined discrepancies between patients and physicians in their assessments of the patient's physical functioning. One recent study reports such discrepancies between rheumatologists and 41% of their rheumatoid arthritis patients. This article reports data replicating that study and examining the relationships between such discrepancies and a number of other variables. METHODS: This is a longitudinal study of 158 patients with rheumatoid arthritis who were interviewed 4 times over a 2-year period and who reported their levels of physical functioning on the Arthritis Impact Measurement Scales. At the time of the fourth interview, the rheumatologists rated each patient's physical functioning on the revised criteria published by the American College of Rheumatology. RESULTS: Rheumatologists' assessments of their patients' physical functioning were discrepant with the patient's assessment for 35% of these patients. Twenty-seven patients were rated as worse than they rated themselves and 28 were rated as better. There were no differences between the concordant and the two discrepant groups in demographic or health status characteristics.

Is the relationship between supine blood pressure and postural changes in blood pressure artefactual? Results from the Kuopio Ischemic Heart Disease Risk Factor Study.

OBJECTIVE: To investigate whether there is an association between initial supine blood pressure and postural changes in blood pressure (standing minus supine blood pressure). METHODS: Using data from the Kuopio (Finland) Ischemic Heart Disease Risk Factor Study (KIHD), we simulated the problem and found the suggested solution based on the work of Blomqvist. We then applied the Blomqvist correction to the KIHD data with real measurement errors. RESULTS: The observed regression slope was substantially reduced, indicating that there is no relationship between the initial blood pressure and the postural change in blood pressure. CONCLUSION: Only the broad application of the method of Blomqvist to other data sets will determine the generalizability of the present finding that initial blood pressure is unrelated to the postural change in blood pressure.

Guidelines for clinical trials in systemic sclerosis (scleroderma). I. Disease-modifying interventions. The American College of Rheumatology Committee on Design and Outcomes in Clinical Trials in Systemic Sclerosis.

OBJECTIVE: To develop guidelines for therapeutic trials designed to improve the overall course of systemic sclerosis (SSc), that is, to reduce the development of significant organ damage or death. METHODS: A committee developed general guidelines for patient inclusion and exclusion criteria, randomization, blinding of patients and physicians, controls, duration of the trial, investigator training, responses, samples size, study dropouts, statistical analyses, data management, and safety monitoring. Delphi and nominal group techniques were used. RESULTS: Briefly, patients with diffuse cutaneous SSc of less than 24 months' duration should be included because they are at greatest risk for the development of severe organ damage and death. Patients should be excluded if they have other connective tissue diseases, SSc-like illnesses related to exposures or ingestions, severe existing internal organ damage, an unacceptable risk of side effects, or concurrent therapies that might independently influence the outcome. Randomized, double-blind, placebo-controlled trials are preferred. The treatment and followup period must be long enough to permit observation of any disease modification, which is likely to require 18-36 months, unless an extraordinarily effective therapy is identified. Responses selected should be quantitative, consistently and accurately reflect activity of SSc in major target organs (not solely the skin), be sensitive to change, and be standardized, with limited variability. An example of a set of responses is given. Surrogate responses are desirable, but none have been validated as correlating with organ damage. CONCLUSION: Guidelines have been established for trials of disease-modifying interventions in SSc. These guidelines will need to be altered as additional information becomes available. Any given protocol will be individualized based on the nature of the intervention and objectives of the study. Nonetheless, each study team should develop a protocol that meets the spirit of these guidelines.

Genital tract abnormalities and female sexual function impairment in systemic sclerosis.

OBJECTIVE: Our purpose was to determine the involvement of the female genital tract and its functional consequences on menstrual and sexual aspects in systemic sclerosis. STUDY DESIGN: Sixty women with systemic sclerosis and 23 age- and disease duration-matched women with either rheumatoid arthritis or systemic lupus erythematosus were surveyed with a comprehensive questionnaire addressing problems before and after disease onset. Fourteen systemic sclerosis patients also had gynecologic evaluations. RESULTS: Vaginal dryness (71%), ulcerations (23%), and dyspareunia (56%) were significantly more frequent in patients with systemic sclerosis after disease onset than before and also in comparison with control subjects. Vaginal tightness and constricted introitus were present in 5 of 60 systemic sclerosis patients. More than half of systemic sclerosis patients reported a decrease in the number (p = 0.04) and intensity (p = 0.02) of orgasms, compared to < 20% of control subjects. The desire and frequency of coitus and the sexual satisfaction index were impaired equally in each group. Skin tightness, reflux-heartburn, and muscle weakness adversely affected sexual relations more in systemic sclerosis than in control subjects. On gynecologic examination 5 of 11 systemic sclerosis patients had small-sized uteri, and 3 of them had early menopause at 29, 38, and 43 years. Seven of 16 (44%) women with systemic sclerosis, compared with 6% of normal women in the United States, attained natural menopause before age 45. CONCLUSIONS: Although impairment in various indexes of sexual function occurs in a number of autoimmune diseases, decreased orgasmic function appears to be limited to systemic sclerosis. Vaginal involvement and other systemic sclerosis-related systemic symptoms adversely influence sexual relations. Menstrual abnormalities, including early menopause, affect many patients. Genital tract involvement occurs in a substantial proportion. Prospective longitudinal studies are warranted.