Variation of a test’s sensitivity and specificity with disease prevalence

Background:

Anecdotal evidence suggests that the sensitivity and specificity of a diagnostic test may vary with disease prevalence. Our objective was to investigate the associations between disease prevalence and test sensitivity and specificity using studies of diagnostic accuracy.

Methods:

We used data from 23 meta-analyses, each of which included 10–39 studies (416 total). The median prevalence per review ranged from 1% to 77%. We evaluated the effects of prevalence on sensitivity and specificity using a bivariate random-effects model for each meta-analysis, with prevalence as a covariate. We estimated the overall effect of prevalence by pooling the effects using the inverse variance method.

Results:

Within a given review, a change in prevalence from the lowest to highest value resulted in a corresponding change in sensitivity or specificity from 0 to 40 percentage points. This effect was statistically significant (p < 0.05) for either sensitivity or specificity in 8 meta-analyses (35%). Overall, specificity tended to be lower with higher disease prevalence; there was no such systematic effect for sensitivity.

Interpretation:

The sensitivity and specificity of a test often vary with disease prevalence; this effect is likely to be the result of mechanisms, such as patient spectrum, that affect prevalence, sensitivity and specificity. Because it may be difficult to identify such mechanisms, clinicians should use prevalence as a guide when selecting studies that most closely match their situation.Diagnostic accuracy plays a central role in the evaluation of medical diagnostic tests. Test accuracy may be expressed as sensitivity and specificity, as positive and negative predictive values or as positive and negative likelihood ratios.¹ Some feel that the positive and negative predictive values of a test are more clinically relevant measures than sensitivity and specificity. However, predictive values directly depend on disease prevalence and can therefore not directly be translated from one situation to another.² In contrast, a test’s sensitivity and specificity are commonly believed not to vary with disease prevalence.^3–5Stability of sensitivity and specificity is an assumption that underlies the use of Bayes theorem in clinical diagnosis. Bayes theorem can be applied in clinical practice by using the likelihood ratio of a test and the probability of the disease before the test was done (pretest probability) to estimate the probability of disease after the test was done.² Because likelihood ratios are a function of sensitivity and specificity, it is assumed that the likelihood ratios also remain the same when prevalence varies.A number of studies have shown that sensitivity and specificity may not be as stable as thought.^6–10 We previously summarized the possible mechanisms through which differences in disease prevalence may lead to changes in a test’s sensitivity and specificity.¹⁰ Prevalence affects diagnostic accuracy because of clinical variability or through artifactual differences, as described in the theoretical framework in 6^,7 Artifactual differences can result from using additional exclusion criteria, verification bias or an imperfect reference standard. For example, using an imperfect reference standard may lead to an underestimate of diagnostic accuracy, but as prevalence increases, the extent to which this happens will vary.^8,9

Table 1:

Theoretical framework of how disease prevalence and test accuracy may be related¹⁰

Estimation and test for linkage between markers: a comparison of lod score and χ 2 test in a linkage study of maritime pine (Pinus pinaster Ait.)

The first step in the construction of a linkage map involves the estimation and test for linkage between all possible pairs of markers. The lod score method is used in many linkage studies for the latter purpose. In contrast with classical statistical tests, this method does not rely on the choice of a first-type error level. We thus provide a comparison between the lod score and a ² test on linkage data from a gymnosperm, the maritime pine. The lod score appears to be a very conservative test with the usual thresholds. Its severity depends on the type of data used.     

Development and application of a two-tier multiple-choice diagnostic test for high school students’ understanding of cell division and reproduction

Department of Physiological Botany, Uppsala University, Uppsala, Sweden Hydrogen gas is regarded as a potential candidate for a future energy economy. Research and development in the field of hydrogen energy is greatly encouraged on all continents. A wide range of microorganisms are able to produce hydrogen gas, among them photosynthetically active organisms that use light as their sole energy source. These organisms are good candidates for the photobiological production of hydrogen gas. Green algae are of particular interest since they are capable of splitting water during photosynthesis and of releasing hydrogen gas under certain conditions. This article describes a small bioreactor that can be run in the classroom and used to demonstrate the concept of photohydrogen production.     

Convergent and construct validity and test–retest reliability of the Caen Chronotype Questionnaire in six languages

ABSTRACT

Chronotype questionnaires provide a simple and time-effective approach to assessing individual differences in circadian variations. Chronotype questionnaires traditionally focused on one dimension of chronotype, namely its orientation along a continuum of morningness and eveningness. The Caen Chronotype Questionnaire (CCQ) was developed to assess an additional dimension of chronotype that captures the extent to which individual functioning varies during the day (amplitude). The aim of this study was to provide a multilanguage validation of the CCQ in six world regions (Arabic, Dutch, German, Italian, Portuguese and Spanish). At Time 1, a total of 2788 participants agreed to take part in the study (Arabic, n = 731; Dutch, n = 538; German, n = 329; Italian, n = 473; Portuguese, n = 361; Spanish, n = 356). Participants completed an assessment of the CCQ together with the Morningness-Eveningness Questionnaire (MEQ; Horne & Ostberg 1976) as well as questions related to factors theoretically related to chronotype (age, shift work, physical activity, sleep parameters and coffee consumption). One month later, participants again completed the CCQ. Results showed that the two-factor structure (morningness-eveningness and amplitude) of the CCQ could be replicated in all six languages. However, measurement invariance could not be assumed regarding the factor loadings across languages, meaning that items loaded more on their factors in some translations than in others. Test–retest reliability of the CCQ ranged from unacceptable (German version) to excellent (Dutch, Portuguese). Convergent validity was established through small–medium effect size correlations between the morningness-eveningness dimension of the CCQ and the MEQ. Taken together, our findings generally support the use of the translated versions of the CCQ. Further validation work on the CCQ is required including convergent validation against physiological markers of sleep, health and well-being.     

Morphological evolution and changes in foraging behaviour of island and mainland populations of Blue Tit (Parus caeruleus) — a test of convergence and ecomorphological hypotheses

Summary We study the leg morphology and feeding postures of two subspecies of the Blue Tit (Parus caeruleus; Tenerife island and the Iberian Peninsula) and the Coal Tit (Parus ater; Iberian Peninsula). We search for evidence supporting the hypothesis of convergent evolution in morphological and ecological traits and we discuss the role of ecomorphological hypotheses as predictors of foraging differences at the intraspecific level. To overcome the problems introduced by environmental characteristics not related to locomotion and competition, we make observations under controlled situations to manage food quality and food access. We determine that the island Blue Tit has a longer tarsometatarsus, larger foot span and a more proximal insertion of the tibialis cranialis muscle (flexor of the tarsometatarsus) than the mainland Blue Tit. These morphological differences are consistent with the more frequent use of hanging and clinging head-up postures by the Iberian Blue Tit. Several ecomorphological hypotheses obtained at the interspecific level with other taxa, have proved to be of high predictive value for explaining ecological differences considering morphological evolution. The Tenerife Blue Tit and the Iberian Coal Tit clearly show close convergence in both feeding postures and leg structure, although some differences in morphology were found between these two species. Convergence in foraging methods between the island Blue Tit and the mainland Coal Tit can be explained without considering current interspecific competition as a determinant of niche space.     

Control of allergic rhinitis and asthma test – a formal approach to the development of a measuring tool

Background

The concurrent management of allergic rhinitis and asthma (ARA) has been recommended by Allergic Rhinitis and its Impact on Asthma (ARIA) guidelines. However, a tool capable of assessing simultaneously the control of upper and lower airways diseases is lacking.

Aim

To describe the studies conducted to design the control of ARA test (CARAT) questionnaire.

Methods

We performed a literature review to generate a list of potentially important items for the assessment of control of ARA. A formal consensus development process, that used an innovative web-based application, was designed – 111 experts in ARA and 60 patients participated. At the final consensus meeting, 25 primary and secondary care physicians formulated the questions and response options. A qualitative feasibility study (n = 31 patients) was conducted to evaluate the comprehensibility of the questionnaire while testing two different designs.

Results

Thirty-four potentially important items were identified. All the steps of the consensus process were completed in 2.5 months. The opinions of experts and patients lead to the formulation of 17 questions. At the feasibility study the instructions and wording problems were corrected and a semi-tabular format was chosen.

Conclusion

A tool to measure the control of allergic rhinitis and asthma was developed using a comprehensive set of methodological steps ensuring the design quality and the face and content validity. Additional validation studies to assess the psychometric properties of the questionnaire have started.     

A test of time–place learning in single and grouped great sturgeon Huso huso

Although there exists a large body of knowledge on the ecology, physiology, and nutrition of great sturgeon, time–place learning has thus far not been examined in this species and other sturgeons. The aim of the present study is to evaluate whether individual or grouped juvenile great sturgeons are able to associate time and place of food delivery when examined under light and dark conditions. During a 30-day period, food was made available to the sturgeons twice a day or twice a night, in the first time in one side of the tank and in the second time in the opposite side of the tank. Food was withheld on test days to assess time–place learning by the fish. The results of the present study indicate that, at both the group and individual levels, great sturgeons are not able to clearly demonstrate time–place learning under the conditions examined.     

Evaluation of on-farm methods for testing the human–animal relationship in dairy herds with cubicle loose housing systems—test–retest and inter-observer reliability and consistency to familiarity of test person

The human–animal relationship is an important factor when considering animal welfare at herd level. In the present study, two behavioural tests for the on-farm assessment of the human–animal relationship at herd level of dairy cows housed in loose housing cubicle systems were evaluated with respect to inter-observer reliability, test–retest reliability, effect of familiarity of test person as well as inter-correlation of the two tests. In a voluntary animal approach (VAA) test, the number of cows and the latencies to approach and touch a stationary test person was measured. In an avoidance (AV) test, the cows’ avoidance reactions to an approaching test person were categorised. A first study was carried out in 12 commercial Danish and Austrian dairy farms. On each farm, both behaviour tests were carried out on the same day and repeated within 4–5 days. For each test, cows were tested by both an unfamiliar and a familiar test person (the stock-person) and two observers simultaneously registered the animals’ test responses. The inter-observer reliability of both behavioural tests was found to be high (VAA: 2.5-m approach r=0.98 (P<0.001) and touch r=0.97 (P<0.001); AV: Kappa coefficient_weighted=0.886 (prevalence index for flight distance≥2 m is 0.636)). The cows at herd level showed shortest latency for touching an unfamiliar test person on the first test day (P=0.006). Further, the AV test had a high test–retest reliability (Kappa coefficient_weighted=0.503 (prevalence index for flight distance≥2 m is −0.660)) and results indicated no effect of familiarity of test person (Kappa coefficient_weighted=0.463 (prevalence index for flight distance≥2 m is −0.677)). In a second study, the correlation between the two behavioural tests (similar measures) was evaluated. On each of 10 commercial Danish dairy farms with loose housing cubicle systems at four repeated sessions, both behaviour tests were carried out on the same day. For each test cows were tested by the stock-person. The VAA and AV tests at herd level were highly correlated (r_s=−0.84; P=0.002).The results suggest that the AV test is valid and applicable for on-farm assessment of the human–animal relationship at herd level. This accounts only partly for the VAA test, which seem to be more unclear regarding motivation for the animals’ approach behaviour.     

How do trees die? A test of the hydraulic failure and carbon starvation hypotheses

Despite decades of research on plant drought tolerance, the physiological mechanisms by which trees succumb to drought are still under debate. We report results from an experiment designed to separate and test the current leading hypotheses of tree mortality. We show that piñon pine (Pinus edulis) trees can die of both hydraulic failure and carbon starvation, and that during drought, the loss of conductivity and carbohydrate reserves can also co‐occur. Hydraulic constraints on plant carbohydrate use determined survival time: turgor loss in the phloem limited access to carbohydrate reserves, but hydraulic control of respiration prolonged survival. Our data also demonstrate that hydraulic failure may be associated with loss of adequate tissue carbohydrate content required for osmoregulation, which then promotes failure to maintain hydraulic integrity.     

Intensity and type of EEG reactions to a hyperventilation test and their relationship with changes in patients’ subjective state of health and some characteristics of the initial EEG pattern

Some EEG characteristics at rest and after a 3-min hyperventilation test (HVT) were studied in 100 adult patients along with the changes in their current subjective state of health, assessed with the help of a special three-stage survey and a questionnaire. Our findings demonstrated a clear correlation between the rate of changes in the current subjective state of health of a patient, the average number of observed symptoms, and the intensity of the EEG reaction. With regard to the methodological aspect, we recorded a difference in the assessment of subjective data obtained through the different types of survey. The frequency of two types of high-amplitude slow waves during the HVT (recorded in 21% of patients), as well as of paroxysmal epileptiform activity, was studied, and their correlation with the clinical manifestations of epilepsy was analyzed. Our study showed that the intensity of the general EEG reaction to the HVT is correlated with the presence of EEG disorders, such as minor bursts in the initial EEG pattern; the probability of high-amplitude slow activity during the HVT is especially high if these minor bursts are recorded at rest. We also noted a certain relationship between the intensity of hyperventilation-related EEG changes (and, therefore, the rate of changes in the current subjective state of health of a patient) and characteristics of the initial pattern, such as α-rhythm disorders and an increase in the intensity of slow or slow and frequent activity at rest.     

Improvement of a β-glucanase activity test by taking into account the batch reactor balance of the test system

Activity tests of enzymes are often applied for determining their concentration. In the easiest case, just one product concentration is measured after a given time. This often leads to nonlinear dependences of the apparent activity with enzyme protein concentration. A general solution of this problem consists in using the balance equation of the assay system, which commonly represents a batch reactor. Here, the balance equation of the batch for a general Michaelis Menten-type reaction kinetics is used as the calibration function. The correlation of the apparent activity and enzyme concentration was established by capturing the enzyme by means of metal chelate interaction owing to a hexahistidine tag attached to the β-glucanase.     

Development and validation of a serological potency test for the release of Leptospira vaccines – Requirements in the European Union

Both European Pharmacopoeia Monograph 01/2008:0447 “Canine Leptospirosis vaccine (inactivated)” and the more recent Monograph 01/2008:1939 “Bovine Leptospirosis vaccine (inactivated)” explicitly allow for a sero-response test to assess batch potency. Test setup and requirements for in vivo and in vitro validation are described. Furthermore, the two main strategies to assess batch potency and their specific demands are addressed.     

Effect of the insecticide Tanrec® on reproduction and vital activity of Daphnia magna Straus in a 15-day test

The effect of the insecticide Tanrec® at concentrations of 3.0 × 10^?7, 3.0 × 10^?2, and 3.0 × 10^?1 mg/L (as of imidacloprid) on Daphnia magna Straus has been studied. An acute toxic effect of this insecticide at a concentration of 3.0 × 10^?1 mg/L and a depressive effect at concentrations of 3.0 × 10^?2 mg/L and 3.0 × 10^?7 have been revealed. A damaging effect of Tanrec was revealed during the stage of early development of studied crustaceans. This effect was manifested in the inhibition of the growth of oocytes, abnormal functioning of the intestine, retardation of body growth, and pathological changes in tissues.     

Identification and Characteristics of a Novel E1 Like Gene nUBE1l in Human test

The selective degradation of many short一lived or abnormal Proteins in eukaryotic cells 15 carried out by the ubiquitin system.In this Pathway,Proteins are targeted for degradation by covalent ligation to ubiquitin,a highly conserved 76一resid…     

Effects of propagule type on genetic parameters of wood density and growth in a loblolly pine progeny test at ages 10 and 11?years

Nine full-sib families of loblolly pine (Pinus taeda L.) were produced by a 3 × 3 factorial mating design. Rooted cuttings and seedlings of full-sib families were tested together in two field locations. Twelve-millimeter wood increment cores were collected from 10- and 11-year-old test trees. On each of the two sites, there were six blocks and a split-plot design, with propagule type as the whole plot and family as the sub-plot. In addition to the collection of wood samples, height and diameter of 1,600 trees were measured. No significant differences were found between cuttings and seedlings for wood density and growth traits. Significant family variation was found for growth and wood density. Genetic parameters estimated for wood density and growth traits using seedlings and rooted cuttings showed that individual-tree and family heritability estimates from rooted cuttings were similar to or higher than those from seedlings for all traits. Half-sib breeding values for parents were highly correlated based on seedling and rooted cutting estimates for height (0.95) and wood density (0.99) but not for diameter (0.56), which suggests that wood density and height breeding value estimates from rooted cuttings in clonal progeny tests can be estimated by traditional seedling tests, but not for tree diameter.     

Range-body mass interactions of a northern ungulate – a test of hypothesis

Summer diet, summer temperature, length of the growth season and animal density appeared to best explain annual and regional differences in calf and yearling body mass in moose from southeastern Norway. In general animals inhabiting steep, alpine landscapes had less body mass than animals using flat, low-altitude habitats. Autumn body mass of calves and yearlings decreased with increasing snow depth during the preceding winter and spring. However, calf body mass was more influenced by the summer range and less by the winter range than was body mass of yearlings. There was no indication that the effect of snow depth on autumn body mass was greater in moose living on poor than on good summer ranges. Body mass decreased with increasing competition for summer forage, while the winter range mainly had an density-independent effect. Habitat quality, expressed as regression lines between calf and yearling body mass and animal density (hunting yield), differed between regions. On ranges of medium and high altitude where birch (Betula spp.) rowan (Sorbus aucuparia) and bilberry (Vaccinium myrtillus) dominated moose summer diet, body mass decreased at a rapid rate with increasing animal density. Body mass decreased at a slower rate at low-altitude ranges and at high-altitude ranges where willow (Salix spp.) and forbs dominated the diet. Body mass of lactating cows decreased with increasing animal density, but animal density did not affect body mass of non-lactating cows. There was no indication that the decrease in autumn body mass with increasing moose density over the last 25 years has caused a decrease in animal condition (ability to survive the winter). The results are discussed in relation to the effect of summer and winter range on population regulation in moose. It is concluded that a density-dependent effect is apparent on the summer range even at low and intermediate population densities. On the winter range, on the other hand, density-dependence is likely to occur only at high levels of population density. Received: 4 February 1997 / Accepted: 1 February 1999