Small proportions: what to report for confidence intervals?
Purpose It is generally agreed that a confidence interval (CI) is usually more informative than a point estimate or p‐value, but we rarely encounter small proportions with CI in the pharmacoepidemiological literature. When a CI is given it is sporadically reported, how it was calculated. This incorr...
Gespeichert in:
| Veröffentlicht in: | Pharmacoepidemiology and drug safety 2005-04, Vol.14 (4), p.239-247 |
|---|---|
| Hauptverfasser: | , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 247 |
|---|---|
| container_issue | 4 |
| container_start_page | 239 |
| container_title | Pharmacoepidemiology and drug safety |
| container_volume | 14 |
| creator | Tobi, Hilde van den Berg, Paul B. de Jong-van den Berg, Lolkje TW |
| description | Purpose
It is generally agreed that a confidence interval (CI) is usually more informative than a point estimate or p‐value, but we rarely encounter small proportions with CI in the pharmacoepidemiological literature. When a CI is given it is sporadically reported, how it was calculated. This incorrectly suggests one single method to calculate CIs. To identify the method best suited for small proportions, seven approximate methods and the Clopper–Pearson Exact method to calculate CIs were compared.
Methods
In a simulation study for 90‐, 95‐ and 99%CIs, with sample size 1000 and proportions ranging from 0.001 to 0.01, were evaluated systematically. Main quality criteria were coverage and interval width. The methods are illustrated using data from pharmacoepidemiology studies.
Results
Simulations showed that standard Wald methods have insufficient coverage probability regardless of how the desired coverage is perceived. Overall, the Exact method and the Score method with continuity correction (CC) performed best. Real life examples showed the methods to yield different results too.
Conclusions
For CIs for small proportions (π ≤ 0.01), the use of the Exact method and the Score method with CC are advocated based on this study. Copyright © 2005 John Wiley & Sons, Ltd. |
| doi_str_mv | 10.1002/pds.1081 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_67557348</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>67557348</sourcerecordid><originalsourceid>FETCH-LOGICAL-c3571-96e1c1784f837566111f2a29bd16185595bc47a7824ce919365351224be4eaba3</originalsourceid><addsrcrecordid>eNp1kMtOwzAQRS0EoqUg8QUoK8QmEMeP2LBAqECLqHgVxNJyHEcEkjjYKaV_j6NGsGI1o5mjO3cuAPswOoZRFJ80mfMNgxtgCCPOQ0hIstn1BIWMUD4AO869R5HfcbwNBpAkkCOCh-BsXsmyDBprGmPbwtTuNFi-yTZoTWB1NwtyYwNl6rzIdK10UNSttl-ydOe7YCv3Ve_1dQRerq-ex9Nwdj-5GV_MQoX8nZBTDRVMGM4ZSgilEMI8ljFPM0ghI4STVOFEJizGSnPvixJEYBzjVGMtU4lG4HCt611-LrRrRVU4pctS1tosnKCJfxdh5sGjNaiscc7qXDS2qKRdCRiJLijhgxJdUB496DUXaaWzP7BPxgPhGlgWpV79KyQeLue9YM8XrtXfv7y0H96f_1u83k3EI-NPs-ktEhT9AHF5fyc</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>67557348</pqid></control><display><type>article</type><title>Small proportions: what to report for confidence intervals?</title><source>MEDLINE</source><source>Wiley Online Library Journals Frontfile Complete</source><creator>Tobi, Hilde ; van den Berg, Paul B. ; de Jong-van den Berg, Lolkje TW</creator><creatorcontrib>Tobi, Hilde ; van den Berg, Paul B. ; de Jong-van den Berg, Lolkje TW</creatorcontrib><description>Purpose
It is generally agreed that a confidence interval (CI) is usually more informative than a point estimate or p‐value, but we rarely encounter small proportions with CI in the pharmacoepidemiological literature. When a CI is given it is sporadically reported, how it was calculated. This incorrectly suggests one single method to calculate CIs. To identify the method best suited for small proportions, seven approximate methods and the Clopper–Pearson Exact method to calculate CIs were compared.
Methods
In a simulation study for 90‐, 95‐ and 99%CIs, with sample size 1000 and proportions ranging from 0.001 to 0.01, were evaluated systematically. Main quality criteria were coverage and interval width. The methods are illustrated using data from pharmacoepidemiology studies.
Results
Simulations showed that standard Wald methods have insufficient coverage probability regardless of how the desired coverage is perceived. Overall, the Exact method and the Score method with continuity correction (CC) performed best. Real life examples showed the methods to yield different results too.
Conclusions
For CIs for small proportions (π ≤ 0.01), the use of the Exact method and the Score method with CC are advocated based on this study. Copyright © 2005 John Wiley & Sons, Ltd.</description><identifier>ISSN: 1053-8569</identifier><identifier>EISSN: 1099-1557</identifier><identifier>DOI: 10.1002/pds.1081</identifier><identifier>PMID: 15719354</identifier><language>eng</language><publisher>Chichester, UK: John Wiley & Sons, Ltd</publisher><subject>binomial proportion ; Confidence Intervals ; Epidemiologic Factors ; Models, Statistical ; Sample Size ; simulation study</subject><ispartof>Pharmacoepidemiology and drug safety, 2005-04, Vol.14 (4), p.239-247</ispartof><rights>Copyright © 2005 John Wiley & Sons, Ltd.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c3571-96e1c1784f837566111f2a29bd16185595bc47a7824ce919365351224be4eaba3</citedby><cites>FETCH-LOGICAL-c3571-96e1c1784f837566111f2a29bd16185595bc47a7824ce919365351224be4eaba3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://onlinelibrary.wiley.com/doi/pdf/10.1002%2Fpds.1081$$EPDF$$P50$$Gwiley$$H</linktopdf><linktohtml>$$Uhttps://onlinelibrary.wiley.com/doi/full/10.1002%2Fpds.1081$$EHTML$$P50$$Gwiley$$H</linktohtml><link.rule.ids>314,776,780,1411,27901,27902,45550,45551</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/15719354$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Tobi, Hilde</creatorcontrib><creatorcontrib>van den Berg, Paul B.</creatorcontrib><creatorcontrib>de Jong-van den Berg, Lolkje TW</creatorcontrib><title>Small proportions: what to report for confidence intervals?</title><title>Pharmacoepidemiology and drug safety</title><addtitle>Pharmacoepidem. Drug Safe</addtitle><description>Purpose
It is generally agreed that a confidence interval (CI) is usually more informative than a point estimate or p‐value, but we rarely encounter small proportions with CI in the pharmacoepidemiological literature. When a CI is given it is sporadically reported, how it was calculated. This incorrectly suggests one single method to calculate CIs. To identify the method best suited for small proportions, seven approximate methods and the Clopper–Pearson Exact method to calculate CIs were compared.
Methods
In a simulation study for 90‐, 95‐ and 99%CIs, with sample size 1000 and proportions ranging from 0.001 to 0.01, were evaluated systematically. Main quality criteria were coverage and interval width. The methods are illustrated using data from pharmacoepidemiology studies.
Results
Simulations showed that standard Wald methods have insufficient coverage probability regardless of how the desired coverage is perceived. Overall, the Exact method and the Score method with continuity correction (CC) performed best. Real life examples showed the methods to yield different results too.
Conclusions
For CIs for small proportions (π ≤ 0.01), the use of the Exact method and the Score method with CC are advocated based on this study. Copyright © 2005 John Wiley & Sons, Ltd.</description><subject>binomial proportion</subject><subject>Confidence Intervals</subject><subject>Epidemiologic Factors</subject><subject>Models, Statistical</subject><subject>Sample Size</subject><subject>simulation study</subject><issn>1053-8569</issn><issn>1099-1557</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2005</creationdate><recordtype>article</recordtype><sourceid>EIF</sourceid><recordid>eNp1kMtOwzAQRS0EoqUg8QUoK8QmEMeP2LBAqECLqHgVxNJyHEcEkjjYKaV_j6NGsGI1o5mjO3cuAPswOoZRFJ80mfMNgxtgCCPOQ0hIstn1BIWMUD4AO869R5HfcbwNBpAkkCOCh-BsXsmyDBprGmPbwtTuNFi-yTZoTWB1NwtyYwNl6rzIdK10UNSttl-ydOe7YCv3Ve_1dQRerq-ex9Nwdj-5GV_MQoX8nZBTDRVMGM4ZSgilEMI8ljFPM0ghI4STVOFEJizGSnPvixJEYBzjVGMtU4lG4HCt611-LrRrRVU4pctS1tosnKCJfxdh5sGjNaiscc7qXDS2qKRdCRiJLijhgxJdUB496DUXaaWzP7BPxgPhGlgWpV79KyQeLue9YM8XrtXfv7y0H96f_1u83k3EI-NPs-ktEhT9AHF5fyc</recordid><startdate>200504</startdate><enddate>200504</enddate><creator>Tobi, Hilde</creator><creator>van den Berg, Paul B.</creator><creator>de Jong-van den Berg, Lolkje TW</creator><general>John Wiley & Sons, Ltd</general><scope>BSCLL</scope><scope>CGR</scope><scope>CUY</scope><scope>CVF</scope><scope>ECM</scope><scope>EIF</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope></search><sort><creationdate>200504</creationdate><title>Small proportions: what to report for confidence intervals?</title><author>Tobi, Hilde ; van den Berg, Paul B. ; de Jong-van den Berg, Lolkje TW</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c3571-96e1c1784f837566111f2a29bd16185595bc47a7824ce919365351224be4eaba3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2005</creationdate><topic>binomial proportion</topic><topic>Confidence Intervals</topic><topic>Epidemiologic Factors</topic><topic>Models, Statistical</topic><topic>Sample Size</topic><topic>simulation study</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Tobi, Hilde</creatorcontrib><creatorcontrib>van den Berg, Paul B.</creatorcontrib><creatorcontrib>de Jong-van den Berg, Lolkje TW</creatorcontrib><collection>Istex</collection><collection>Medline</collection><collection>MEDLINE</collection><collection>MEDLINE (Ovid)</collection><collection>MEDLINE</collection><collection>MEDLINE</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Pharmacoepidemiology and drug safety</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Tobi, Hilde</au><au>van den Berg, Paul B.</au><au>de Jong-van den Berg, Lolkje TW</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Small proportions: what to report for confidence intervals?</atitle><jtitle>Pharmacoepidemiology and drug safety</jtitle><addtitle>Pharmacoepidem. Drug Safe</addtitle><date>2005-04</date><risdate>2005</risdate><volume>14</volume><issue>4</issue><spage>239</spage><epage>247</epage><pages>239-247</pages><issn>1053-8569</issn><eissn>1099-1557</eissn><abstract>Purpose
It is generally agreed that a confidence interval (CI) is usually more informative than a point estimate or p‐value, but we rarely encounter small proportions with CI in the pharmacoepidemiological literature. When a CI is given it is sporadically reported, how it was calculated. This incorrectly suggests one single method to calculate CIs. To identify the method best suited for small proportions, seven approximate methods and the Clopper–Pearson Exact method to calculate CIs were compared.
Methods
In a simulation study for 90‐, 95‐ and 99%CIs, with sample size 1000 and proportions ranging from 0.001 to 0.01, were evaluated systematically. Main quality criteria were coverage and interval width. The methods are illustrated using data from pharmacoepidemiology studies.
Results
Simulations showed that standard Wald methods have insufficient coverage probability regardless of how the desired coverage is perceived. Overall, the Exact method and the Score method with continuity correction (CC) performed best. Real life examples showed the methods to yield different results too.
Conclusions
For CIs for small proportions (π ≤ 0.01), the use of the Exact method and the Score method with CC are advocated based on this study. Copyright © 2005 John Wiley & Sons, Ltd.</abstract><cop>Chichester, UK</cop><pub>John Wiley & Sons, Ltd</pub><pmid>15719354</pmid><doi>10.1002/pds.1081</doi><tpages>9</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1053-8569 |
| ispartof | Pharmacoepidemiology and drug safety, 2005-04, Vol.14 (4), p.239-247 |
| issn | 1053-8569 1099-1557 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_67557348 |
| source | MEDLINE; Wiley Online Library Journals Frontfile Complete |
| subjects | binomial proportion Confidence Intervals Epidemiologic Factors Models, Statistical Sample Size simulation study |
| title | Small proportions: what to report for confidence intervals? |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-12T12%3A03%3A10IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Small%20proportions:%20what%20to%20report%20for%20confidence%20intervals?&rft.jtitle=Pharmacoepidemiology%20and%20drug%20safety&rft.au=Tobi,%20Hilde&rft.date=2005-04&rft.volume=14&rft.issue=4&rft.spage=239&rft.epage=247&rft.pages=239-247&rft.issn=1053-8569&rft.eissn=1099-1557&rft_id=info:doi/10.1002/pds.1081&rft_dat=%3Cproquest_cross%3E67557348%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=67557348&rft_id=info:pmid/15719354&rfr_iscdi=true |