Linked Life Data

10117988

Source:http://linkedlifedata.com/resource/pubmed/id/10117988

Statements in which the resource exists as a subject.
PredicateObject
rdf:type
lifeskim:mentions
pubmed:issue
1
pubmed:dateCreated
1992-6-23
pubmed:abstractText
Health interview surveys have been widely used to measure morbidity in developing countries, particularly for infectious diseases. Structured questionnaires using algorithms which derive sign/symptom-based diagnoses seem to be the most reliable but there have been few studies to validate them. The purpose of validation is to evaluate the sensitivity and specificity of brief algorithms (combinations of signs/symptoms) which can then be used for the rapid assessment of community health problems. Validation requires a comparison with an external standard such as physician or serological diagnoses. There are several potential pitfalls in assessing validity, such as selection bias, differences in populations and the pattern of diseases in study populations compared to the community. Validation studies conducted in the community may overcome bias caused by case selection. Health centre derived estimates can be adjusted and applied to the community with caution. Further study is needed to validate algorithms for important diseases in different cultural settings. Community-based studies need to be conducted, and the utility of derived algorithms for tracking disease frequency explored further.
pubmed:keyword
http://linkedlifedata.com/resource/pubmed/keyword/Child Health, http://linkedlifedata.com/resource/pubmed/keyword/Communication, http://linkedlifedata.com/resource/pubmed/keyword/Critique, http://linkedlifedata.com/resource/pubmed/keyword/Data Collection, http://linkedlifedata.com/resource/pubmed/keyword/Developing Countries, http://linkedlifedata.com/resource/pubmed/keyword/Diseases, http://linkedlifedata.com/resource/pubmed/keyword/Estimation Technics, http://linkedlifedata.com/resource/pubmed/keyword/Health, http://linkedlifedata.com/resource/pubmed/keyword/INTERVIEWS, http://linkedlifedata.com/resource/pubmed/keyword/Indirect Estimation Technics, http://linkedlifedata.com/resource/pubmed/keyword/LANGUAGE, http://linkedlifedata.com/resource/pubmed/keyword/MORBIDITY, http://linkedlifedata.com/resource/pubmed/keyword/Measurement, http://linkedlifedata.com/resource/pubmed/keyword/Prevalence, http://linkedlifedata.com/resource/pubmed/keyword/Questionnaire Design, http://linkedlifedata.com/resource/pubmed/keyword/Research Methodology, http://linkedlifedata.com/resource/pubmed/keyword/Sampling Studies, http://linkedlifedata.com/resource/pubmed/keyword/Signs And Symptoms, http://linkedlifedata.com/resource/pubmed/keyword/Studies, http://linkedlifedata.com/resource/pubmed/keyword/Survey Methodology, http://linkedlifedata.com/resource/pubmed/keyword/Surveys
pubmed:language
eng
pubmed:journal
pubmed:citationSubset
H
pubmed:status
MEDLINE
pubmed:month
Mar
pubmed:issn
0268-1080
pubmed:author
pubmed:issnType
Print
pubmed:volume
7
pubmed:owner
NLM
pubmed:authorsComplete
Y
pubmed:pagination
30-9
pubmed:dateRevised
2007-11-15
pubmed:otherAbstract
PIP: Health interviews are conducted for reasons of rapid assessment, of portraying regional or community differences in health and service utilization, and of assessing the health of the general population. The validation process is discussed as it pertains to the potential biases: sensitivity, specificity, prevalence, and predictive value. A mathematical formula is provided in table form and is expressed as functions of the biases. A comparison of interview-based diagnoses with some external standard directly with medical, serological, or other laboratory diagnoses is required to measure the validity of predefined algorithm or disease reports. A comparison with an epidemiological pattern or outcome of a prevention campaign or controlled intervention study leads to only a general conclusion. The development of an interview instrument for validation is discussed, as well as the factors affecting the validation of the instrument: cost, characteristics of candidate illnesses, linguistics and respondent recognition of the illness, recall of the illness and its signs, imperfect external standards, and selection bias. In the transition to the field, there is further discussion of the choice of algorithm, estimating disease prevalence and predictive value, and adjusting for selection bias. There have been few formal validation studies of sign-based diagnoses derived from structured questionnaires to determine the sensitivity and specificity of diagnostic algorithms. The advantages of validated algorithms are the reliability in calculating community disease prevalence, the provision for rapid assessment and sensitivity to culture-specific settings, and the provision for testing the degree of sign recognition and for determining acceptable recall intervals. The structure and content of the study instrument affect validity. A modular questionnaire is the most efficient. Content should be determined by an exploratory, empirical approach which incorporates biological and cultural factors. This means inclusion of local disease terminology, complete and accurately translated sign lists, and questions on timing and duration of signs. The impact of recognition and recall of signs should be taken into consideration in instrument and study design. Selection bias is caused by a discrepancy between the medical and community setting. Compensation for selection bias can be made studying the impact of demographic characteristics and the distribution of disease in the 2 settings. This increases the practical utility of structured morbidity interviews.
pubmed:meshHeading
pubmed:year
1992
pubmed:articleTitle
The validation of interviews for estimating morbidity.
pubmed:affiliation
Carney Hospital, Boston.
pubmed:publicationType
Journal Article