Cross sectional studies | My Assignment Tutor

See discussions, stats, and author profiles for this publication at: sectional studies: advantages and disadvantagesArticle in BMJ (online) · March 2014DOI: 10.1136/bmj.g2276CITATIONS134READS20,5851 author:Some of the authors of this publication are also working on these related projects:BMJ Statistics Endgames View projectPhilip SedgwickSt George’s, University of London410 PUBLICATIONS 5,865 CITATIONSSEE PROFILEAll content following this page was uploaded by Philip Sedgwick on 07 February 2017.The user has requested enhancement of the downloaded file.STATISTICAL QUESTIONCross sectional studies: advantages and disadvantagesPhilip Sedgwick reader in medical statistics and medical educationCentre for Medical and Healthcare Education, St George’s, University of London, London, UKA cross sectional study design was used to investigate the extentof chronic fatigue and the associated psychosocial exposures ina developing country. The setting was a primary health centrecatchment area in Goa, India. Participants were women aged18-50 years. The primary outcome was reporting of fatigue forat least six months. Data on the primary outcome andpsychosocial exposures were obtained by a structured interview.The psychosocial exposures that were investigated includedmental health and gender disadvantage factors. The presenceof anaemia was determined from a blood sample.1The sampling frame consisted of 8595 eligible women listed infamily health registers. In total, 3000 randomly sampled womenwere invited to participate and 2494 (83%) consented.Recruitment took place from November 2001 to May 2003. Itwas reported that 12.1% (95% confidence interval 10.8 to13.4%) of women experienced chronic fatigue. Thosepsychosocial exposures most strongly associated with chronicfatigue were poor mental health and sexual violence by thehusband.Which of the following statements, if any, are true?a) The study was longitudinal by designb) It was possible to estimate the prevalence of chronicfatigue in women aged 18-50 yearsc) It was possible to estimate the incidence of chronic fatiguein women aged 18-50 yearsd) It can be inferred that chronic fatigue was caused by poormental health and sexual violence by the husbandAnswersStatement b is true, whereas a, c, and d are false.The aim of the study was to investigate the extent of chronicfatigue and the associated psychosocial exposures in adeveloping country. A cross sectional study, which isobservational in design, was undertaken in Goa, India. In anobservational study the investigators do not intervene in anyway but simply record the health, behaviour, attitudes, orlifestyle choices of the study participants. As the name suggests,in a cross sectional study the researchers aim to obtain arepresentative sample by taking a cross section of the population.In the study above, the population was women aged 18-50 yearswho lived in a developing country. The study participants werea random sample of women from the family health registers forthe catchment area in Goa. Other types of sampling could havebeen used to recruit women, including convenience sampling,which has been described in a previous question.2In a cross sectional study, all the measurements for a samplemember are obtained at a single point in time, althoughrecruitment may take place across a longer period of time. Inthe study above, each woman was interviewed once, withrecruitment taking place between November 2001 and May2003.A cross sectional study is not longitudinal by design (a is false).In a longitudinal study, each participant is observed at multipletime points, thereby allowing trends in an outcome to bemonitored over time. Longitudinal studies may be prospectiveor retrospective and observational or experimental in design.The cohort study, described in previous questions,3 4 is anexample of a longitudinal study that is observational in design.Because women in the study above were interviewed only once,it was not possible to describe trends in chronic fatigue.A cross sectional study is particularly suitable for estimatingthe prevalence of a behaviour or disease in a population.Prevalence is the proportion of the population that has thebehaviour or disease. In the study above 12.1% of womenreported chronic fatigue, and this figure is an estimate of theprevalence of chronic fatigue in the population of women aged18-50 years in a developing country (b is true). Incidence is thenumber of new cases of a particular behaviour or disease in thepopulation that occur within a specified period, expressed as aproportion of the number of people in the population duringthat time. It was not possible to estimate the incidence of chronicfatigue in the population of women aged 18-50 years in adeveloping country from the study above (c is false). Womenwere interviewed only once for assessment of chronic fatigue.They were not followed longitudinally to assess whether newcases of chronic fatigue developed. Therefore, the proportionof new cases of chronic fatigue in the sample over a particulartime period could not be [email protected] personal use only: See rights and reprints Subscribe: 2014;348:g2276 doi: 10.1136/bmj.g2276 (Published 26 March 2014) Page 1 of 2EndgamesENDGAMESIncidence and prevalence can be reported as a proportion,percentage, or as a rate per 1000, per 100 000, or per 1 000 000,depending on the rarity of the behaviour or disease. Suchmeasures of the occurrence of a behaviour or disease are usefulwhen allocating resources and planning healthcare services.Cross sectional studies are generally quick, easy, and cheap toperform. They are often based on a questionnaire survey. Therewill be no loss to follow-up because participants are interviewedonly once. However, a cross sectional study may be prone tonon-response bias if participants who consent to take part in thestudy differ from those who do not, resulting in a sample thatis not representative of the population. It is possible to recordexposure to many risk factors and to assess more than oneoutcome in a cross sectional study. However, because data oneach participant are recorded only once it would be difficult toinfer the temporal association between a risk factor and anoutcome. Therefore, only an association, and not causation, canbe inferred from a cross sectional study (d is false). For the studyabove, it is not possible to infer that poor mental health andsexual violence by the husband preceded the onset of chronicfatigue and that they may be risk factors for chronic fatigue. Itcan be inferred only that chronic fatigue was associated withpoor mental health and sexual violence by the husband. Theresults from a cross sectional study may inform the hypothesesfor a more complex investigation, such as a cohort study.Cross sectional studies are sometimes repeated at different timesto assess trends over time. However, caution is needed ifdifferent participants are included at each time point. It may bedifficult to assess whether changes in prevalence reflect a trendor simply differences between different groups of participantssampled from the population.Competing interests: None declared.1 Patel V, Kirkwood BR, Weiss H, Pednekar S, Fernandes J, Pereira B, et al. Chronic fatiguein developing countries: population based survey of women in India. BMJ 2005;330:1190.2 Sedgwick P. Convenience sampling. BMJ 2013;347:f6304.3 Sedgwick P. Prospective cohort studies: advantages and disadvantages. BMJ2013;347:f6726.4 Sedgwick P. Retrospective cohort studies: advantages and disadvantages. BMJ2014;348:g1072.Cite this as: BMJ 2014;348:g2276© BMJ Publishing Group Ltd 2014For personal use only: See rights and reprints Subscribe: 2014;348:g2276 doi: 10.1136/bmj.g2276 (Published 26 March 2014) Page 2 of 2ENDGAMESView publication stats


Leave a Reply

Your email address will not be published. Required fields are marked *