How accurate are subjective judgements of a continuum?

AIMS: To assess the accuracy of subjective judgements of a uni-dimensional continuum. METHODS AND RESULTS: Twenty-four observers were asked to estimate the position of a dot placed between two lines using a 100-point scale. Each observer judged 20 different positions. Ninety-five percent of observations were between -7 and +9 of the true value. The median and the mean value of the estimates for each position were within one or two points of the true value in most cases with a maximum deviation of five. There was perfect correlation between the ranks of the median and the mean and the true value (correlation coefficient +1, P < 0.00001). The kappa values for interobserver agreement varied from 0.98 for two categories to 0.55 for 10 categories and 0.39 for 20 categories. Information transmitted increased from 0.98 bits with two categories to 2.83 bits with 20 categories. CONCLUSIONS: The experiment indicates that subjective judgements can be accurate but it is necessary to use a large number of categories. One hundred-point scales allow confidence intervals to be calculated and the intervals are reduced if the results from several observers are pooled. The mean and the median of the judgements of a large number of observers are close to objective reality and this approach could be used to define a gold standard of rankings of pathological continua such as cervical dyskaryosis or tumour differentiation. There is a reciprocal relationship between kappa values and information transmitted which demonstrates the fallacy of using kappa statistics in studies of continua.     

Did Robert Louis Stevenson have hereditary hemorrhagic telangiectasia?

Chronic illness played a major role in the life and literary success of Robert Louis Stevenson. However, the exact nature of his chronic illness remains unclear. It is possible that Stevenson had hereditary hemorrhagic telangiectasia (Osler‐Rendu‐Weber Syndrome). This would explain his chronic respiratory complaints, recurrent episodes of pulmonary hemorrhage, and his death, at age 44 years, of probable cerebral hemorrhage. It would also explain his mother's hitherto unreported but apparent stroke, at age 38 years. Further support for this hypothesis might come from new details about the health of Stevenson and his relatives or from molecular analysis of tissue specimens remaining from him. Am. J. Med. Genet. 91:62–65, 2000 © 2000 Wiley‐Liss, Inc.     

Facial measurements in clinical genetics: How important are the instruments we use?

Prompted by our finding that a popular compendium of clinical measurements often suggests a transparent ruler as a suitable substitute for anthropometric calipers (which were typically used by the original researchers to collect the normative data), we compared facial measurements taken with a ruler and calipers. Our objectives were to compare facial measurement data taken with these instruments by two classes of observer: expert and inexperienced. Ten facial measurements were repeated on four medically normal women by one expert and one inexperienced observer. Both observers' data showed that the caliper-derived means were usually the larger, but, whereas the expert observer's caliper-derived data typically were the least variable, the novice observer had smaller standard deviations and ranges for the ruler-derived data. Statistically significant differences were found between the ruler- and caliper-derived data from both observers on all four subjects, except for subnasale-pogonion and stomion-pogonion. For the novice observer only, endocanthion-endocanthion, left exocanthion-endocanthion, and alare-alare were also nonsignificant. The calibrations of the sliding caliper and ruler were compared to determine if differences between them could explain the statistical results, but were the same. We concluded that the differences between the caliper- and ruler-derived measurements resulted because the ruler often could not be placed directly on the landmarks, as could the arms of the calipers. We recommend that clinicians interested in taking facial measurements to assess their patients consult the original publications for information on the techniques and instruments used so that reliable comparisons with the normative data can be made. Am. J. Med. Genet. 77:384–390, 1998. © 1998 Wiley-Liss, Inc.     

Did Robert Louis Stevenson have hereditary hemorrhagic telangiectasia?

Chronic illness played a major role in the life and literary success of Robert Louis Stevenson. However, the exact nature of his chronic illness remains unclear. It is possible that Stevenson had hereditary hemorrhagic telangiectasia (Osler-Rendu-Weber Syndrome). This would explain his chronic respiratory complaints, recurrent episodes of pulmonary hemorrhage, and his death, at age 44 years, of probable cerebral hemorrhage. It would also explain his mother's hitherto unreported but apparent stroke, at age 38 years. Further support for this hypothesis might come from new details about the health of Stevenson and his relatives or from molecular analysis of tissue specimens remaining from him.     

Frequency of epistaxis and telangiectasia in patients with hereditary hemorrhagic telangiectasia (HHT) in comparison with the general population: Curaçao diagnostic criteria revisited

PurposeThe Curaçao criteria are well-established diagnostic criteria for hereditary hemorrhagic telangiectasia (HHT), but they lack details regarding a predictive presentation of epistaxis and telangiectasias. This study collects and compares data in HHT and population cohorts to inform the application of these criteria.MethodsIn-person interviews regarding epistaxis and targeted examination for telangiectases in a general population cohort (n = 204) and an HHT cohort (n = 432) were conducted.ResultsFrequency of epistaxis, rather than intensity or duration, was the best discriminator of HHT. A cutoff of ≥4 nosebleeds per year alone yielded a diagnostic sensitivity of 97%, and specificity of 84%. The mean number of telangiectases at the sites investigated was 0.4 in the general population cohort and 26.5 in the HHT cohort. The most distinctive sites for telangiectases in HHT were lips and palmar fingers, whereas telangiectases of the face and dorsum of the hand were comparable in both cohorts.ConclusionWe propose that the Curaçao criteria be modified to include the following cutoffs: (1) epistaxis frequency of ≥4 nosebleeds per year and (2) telangiectasia count of at least 2 in characteristic locations (palmar aspect of fingers, lips, and oral cavity), and that cutaneous telangiectases at other sites not be considered relevant for diagnostic purposes.     

How accurate are patients in reporting their antidepressant treatment history?

BACKGROUND: A patient's report of their antidepressant treatment history is one of the most important pieces of information used in selecting an antidepressant regimen. It is currently unknown how accurate patients are in describing and characterizing their antidepressant treatment history. METHODS: Seventy-three patients receiving treatment for depression at our outpatient psychiatric practice were interviewed by an independent evaluator who was blind to each patient's treatment history. Information was obtained regarding which antidepressant and augmentation regimens patients had undergone, antidepressant doses, duration of trials, and the nature of response to each trial. The results of these interviews were then compared with patients' actual treatment history as elicited from an independent chart review. RESULTS: Patients recalled 85 of the 104 (81.7%) monotherapy trials they had undergone in the past 5 years, but only recalled 12 of 46 (26.1%) augmentation trials (P<0.001). Patients were found to be very reliable in distinguishing between those trials that were of adequate dose and duration and those that were not. Patients were also generally reliable in depicting the quality of response to past trials, though patient report of a past negative trial was significantly more reliable than a report of a past positive trial. The presence of current depressive symptomatology did not adversely affect patients' ability to recall past trials or accurately describe their responses to past regimens. LIMITATIONS: All patients were treated by a single psychiatrist, and all trials occurred within the last 5 years. CONCLUSION: Patients are able to recall the majority of monotherapy trials they have undergone, but have great difficulty remembering when two medications were taken concurrently, i.e. augmentation trials. Patient report appears to be a satisfactory method to obtain information regarding trial adequacy and response in most, but not all instances.     

Human body donation: How informed are the donors?

Deceased human bodies are donated for education and research. Informed consent has become the standard for research on the living. A question could be asked on how informed are the donors and their families about the process before this generous gift is given. The aim of this study is to evaluate and compare the published donation forms used by body donation programs in the United States and assess them according to the guidelines published by the American Association of Clinical Anatomists. The findings of this study shows that the level of information given to donors and families, before consenting to whole body donation, varies greatly throughout the United States. Many of the forms fail to include the recommendations made by professional societies. Additional information needs to be added to whole body donation forms to better inform donors and families about the donation process, what happens to the body, and the final disposition of the bodies once studies are completed. Overall, it was concluded that in some cases consent is being obtained but much more needs to be done before institutions can claim to obtain informed consent.     

How useful are clinical guidelines for the management of obesity in general practice?

Obesity is a major public health issue and numerous clinical guidelines have been published to support management. One of the most comprehensive guidelines on obesity was published by the National Institute for Health and Clinical Excellence (NICE) in 2006 (NICE guideline 43) which aims to offer practical recommendations based on the available evidence and has a strong focus on primary care both in terms of prevention and clinical care. The current article summarises these guidelines in relation to primary care, reports on new evidence and developments since they were published, and critically appraises the usefulness of guidelines for management of obesity.     

How accurate are lockable orthotic knee braces? An objective gait analysis study

There has been an increasing use of orthotic knee braces in the management of knee injuries but, to our knowledge, there is no gait analysis study assessing the accuracy of these braces. Eight healthy male subjects were studied to determine the accuracy of immobilisation or splintage provided by a lockable orthotic knee brace using gait analysis. Six types of immobilisation were studied: locked at 0, 10, 20, 30° and unlocked in an orthotic knee brace, and without a brace. The knee flexion angles measured using the kinematic instruments at 0 and 10° were significantly greater than those set at the knee brace. The knee flexion angle measured using the unlocked knee brace was significantly greater than that measured in the absence of a brace. This study highlights inaccuracies in a knee brace at low knee flexion angles. The higher actual angles alter the biomechanics of the knee joint and result in greater forces across the knee joint and especially the extensor mechanism.     

PTEN immunohistochemistry in endometrial hyperplasia: which are the optimal criteria for the diagnosis of precancer?

Guidelines recommend protein phosphatase and tensin homolog (PTEN) immunohistochemistry for differentiating between benign endometrial hyperplasia (BEH) and atypical endometrial hyperplasia/endometrioid intraepithelial neoplasia (AEH/EIN). However, it is unclear when PTEN expression should be defined as ‘lost’ and thus suggestive of AEH/EIN. We aimed to determine the optimal immunohistochemical criteria to define PTEN loss in endometrial hyperplasia, through a systematic review and meta‐analysis of diagnostic accuracy. Electronic databases were searched for studies assessing immunohistochemical expression of PTEN in both BEH and AEH/EIN specimens. PTEN status (‘loss’ or ‘presence’) was the index test; histological diagnosis (‘AEH/EIN’ or ‘BEH’) was the reference standard. Accuracy was quantified based on the area under the curve (AUC) on summary receiver operating characteristic (SROC) curves, for several different thresholds of PTEN expression. Eighteen studies with 1362 hyperplasias were included. Six different criteria to define PTEN loss were assessed. Low diagnostic accuracy was found for complete loss of expression (AUC = 0.71), presence of any null gland (AUC = 0.63), positive cells <10% (AUC = 0.64), positive cells <50% (AUC = 0.71) and moderate‐to‐null intensity (AUC = 0.64). Barely moderate diagnostic accuracy was only found for the subjective criterion ‘weak‐to‐null intensity’ (AUC = 0.78). Therefore, the clinical usefulness of PTEN immunohistochemistry in this field should be further investigated.     

How valid are the subscales of the multiscore depression inventory?

Investigated subscales of the Multiscore Depression Inventory for construct and concurrent validity. A hierarchical cluster analysis based on the responses of 263 students provided support for the construct validity of Irritability, Energy Level, Guilt, Instrumental Helplessness, Social Introversion, and Cognitive Difficulty. The separation of Low Self-Esteem and Pessimism was supported only partially, while Learned Helplessness and Sad Mood combined in a single cluster. The graph-theoretic advantages of the cluster analysis permitted speculation about the relationship among the subscales. Factoranalytic examination of the same data demonstrated comparable evidence of construct validity for most scales. Concurrent validity was demonstrated on a sample of 89 students, with significant relationships obtained for all subscales with respective criterion measures, with the exception of Learned Helplessness, which had a low, but significant, correlation with its criterion. Discriminant validity was demonstrated by low correlations between the subscales and socially desirable responses.