Introduction - University of Pittsburgh



9144001143000MISSED OPPORTUNITIES: AN ASSESSMENT OF LEGIONELLOSIS SURVEILLANCE IN ALLEGHENY COUNTYbyLindsay HamiltonBachelor of Science in Microbiology, Virginia Polytechnic Institute and State University, 2012Submitted to the Graduate Faculty ofthe Department of Epidemiology in theGraduate School of Public Health in partial fulfillment of the requirements for the degree ofMaster of Public HealthUniversity of Pittsburgh201400MISSED OPPORTUNITIES: AN ASSESSMENT OF LEGIONELLOSIS SURVEILLANCE IN ALLEGHENY COUNTYbyLindsay HamiltonBachelor of Science in Microbiology, Virginia Polytechnic Institute and State University, 2012Submitted to the Graduate Faculty ofthe Department of Epidemiology in theGraduate School of Public Health in partial fulfillment of the requirements for the degree ofMaster of Public HealthUniversity of Pittsburgh2014center301625UNIVERSITY OF PITTSBURGHGRADUATE SCHOOL OF PUBLIC HEALTHThis essay is submittedbyLindsay HamiltononApril 18, 2014and approved byEssay Advisor:Ronald E. Voorhees, MD, MPH______________________________________Professor of Public Health PracticeEpidemiologyGraduate School of Public HealthUniversity of PittsburghEssay Reader:Nicholas G. Castle, PhD______________________________________ProfessorHealth Policy and ManagementGraduate School of Public HealthUniversity of PittsburghEssay Reader:Evelyn O. Talbott, DrPH, MPH______________________________________ProfessorEpidemiologyGraduate School of Public HealthUniversity of Pittsburgh00UNIVERSITY OF PITTSBURGHGRADUATE SCHOOL OF PUBLIC HEALTHThis essay is submittedbyLindsay HamiltononApril 18, 2014and approved byEssay Advisor:Ronald E. Voorhees, MD, MPH______________________________________Professor of Public Health PracticeEpidemiologyGraduate School of Public HealthUniversity of PittsburghEssay Reader:Nicholas G. Castle, PhD______________________________________ProfessorHealth Policy and ManagementGraduate School of Public HealthUniversity of PittsburghEssay Reader:Evelyn O. Talbott, DrPH, MPH______________________________________ProfessorEpidemiologyGraduate School of Public HealthUniversity of Pittsburghcenter4648200Copyright ? by Lindsay Hamilton201400Copyright ? by Lindsay Hamilton2014center-222250Ronald E. Voorhees, MD, MPHMISSED OPPORTUNITIES: AN ASSESSMENT OF LEGIONELLOSIS SURVEILLANCE IN ALLEGHENY COUNTYLindsay Hamilton, MPHUniversity of Pittsburgh, 2014ABSTRACT00Ronald E. Voorhees, MD, MPHMISSED OPPORTUNITIES: AN ASSESSMENT OF LEGIONELLOSIS SURVEILLANCE IN ALLEGHENY COUNTYLindsay Hamilton, MPHUniversity of Pittsburgh, 2014ABSTRACTSince the first outbreak and discovery of the pneumonia-causing bacteria Legionella in 1976 at a hotel convention in Philadelphia, there have been major advances in scientific knowledge about the disease, including the creation and improvements of diagnostic tests. There is still a serious lack of information due to undiagnosed cases, incomplete surveillance of the disease, and a general lack of research on the effectiveness of remediation methods. In partnership with the Allegheny County Health Department, I started a project with the goals of 1) finding odds of having certain risk factors when developing legionellosis, 2) finding the relative risk of having a repeat case in a long-term care facility or SHR after environmental testing, and 3) finding the relative risk of having a repeat case in a long-term care facility after a given remediation recommendation, all using PA-NEDSS elderly and health-care exposed legionellosis cases from 2003-2013. When it became apparent that there was too much missing or misclassified data for the analysis to yield reliable results, the focus of the project shifted and its goals became to 1) assess the amount or proportion of missing data for all variables of interest, 2) analyze the present surveillance system, and 3) make recommendations based on my analysis of the surveillance system. Several important aspects of each case report were analyzed in order to summarize the effect of missing data in the surveillance system. CDC guidelines for evaluating a public health surveillance system were used to summarize problems in the PA-NEDSS system and to make appropriate recommendations. To know how well a surveillance system is working at a local or state health department can significantly contribute to public health knowledge and practice since a surveillance system can directly and indirectly contribute to monitoring health trends, understanding risk factors, and determining the most effective prevention methods for a given population.TABLE OF CONTENTS TOC \o "2-4" \h \z \t "Heading 1,1,Appendix,1,Heading,1" preface PAGEREF _Toc259891589 \h viii1.0Introduction PAGEREF _Toc259891590 \h 11.1Incidence of legionnaire’s disease PAGEREF _Toc259891591 \h 11.2Surveillance of legionnaire’s disease PAGEREF _Toc259891592 \h 41.2.1Evaluating a Surveillance System PAGEREF _Toc259891593 \h 51.3guidelines for legionnaire’s disease PAGEREF _Toc259891594 \h 62.0Risk Factors, Environmental Testing, And REmediation PAGEREF _Toc259891595 \h 92.1Methods PAGEREF _Toc259891596 \h 92.2results PAGEREF _Toc259891597 \h 113.0Analysis of Allegheny COunty Legionellosis surveillance PAGEREF _Toc259891598 \h 133.1methods PAGEREF _Toc259891599 \h 133.2results PAGEREF _Toc259891600 \h 144.0Discussion PAGEREF _Toc259891601 \h 174.1Surveillance System Problems PAGEREF _Toc259891602 \h 184.2Recommendations PAGEREF _Toc259891603 \h 19APPENDIX: TIMEline of risk factor variables PAGEREF _Toc259891604 \h 22bibliography PAGEREF _Toc259891605 \h 23List of tables TOC \h \z \c "Table" Table 1. Summary characteristics for total Allegheny County reported legionellosis cases PAGEREF _Toc259891617 \h 11Table 2. Missing data on variables for health-care exposed cases PAGEREF _Toc259891618 \h 14List of figures TOC \h \z \c "Figure" Figure 1. Annual number and age-adjusted incidence rate per 100,000 population of reported legionella cases in US, reported through the Centers for Disease Control for 1990-2005 PAGEREF _Toc259891627 \h 2Figure 2. Annual reported rates of legionellosis, US and Allegheny County PAGEREF _Toc259891628 \h 2Figure 3. Average age-specific rates of reported legionellosis cases in US (2000-2009) and Allegheny County (2003-2013); Note: different age classes are used for US and Allegheny County Rates PAGEREF _Toc259891629 \h 3Figure 4. Annual age-specific rates of reported legionellosis in Allegheny County from 2003-2013 PAGEREF _Toc259891630 \h 12prefaceI would like to acknowledge the Allegheny County Health Department for the use of its data and services, and especially thank the staff of the Office of Epidemiology and Biostatistics, specifically Lauren Torso. For the purposes of this essay, ‘Legionellosis’ will refer to ‘Legionnaire’s disease’ unless otherwise specified. This is because the national surveillance system for legionellosis does not specify between Pontiac Fever and Legionnaire’s disease cases when referring to legionellosis, and surveillance staff indicate that a minute number of Pontiac Fever cases entered the surveillance system for the 2003-2013 year period discussed ADDIN EN.CITE <EndNote><Cite><Author>Torso</Author><Year>2014</Year><RecNum>22</RecNum><DisplayText>(Torso 2014)</DisplayText><record><rec-number>22</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1397080541">22</key></foreign-keys><ref-type name="Interview">64</ref-type><contributors><authors><author>Lauren Torso</author></authors><secondary-authors><author>Lindsay Hamilton</author></secondary-authors></contributors><titles><title>Problems with PA-NEDSS</title><secondary-title>University of Pittsburgh Graduate School of Public Health</secondary-title><tertiary-title>Allegheny County Health Department</tertiary-title></titles><dates><year>2014</year></dates><urls></urls></record></Cite></EndNote>(Torso 2014).Introduction‘Legionellosis’ consists of ‘Legionnaire’s Disease’, a type of pneumonia caused by the bacteria Legionella that can be naturally found in water, and ‘Pontiac Fever’, a less severe flu-like manifestation ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2013</Year><RecNum>1</RecNum><DisplayText>(CDC 2013)</DisplayText><record><rec-number>1</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395349901">1</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author>CDC</author></authors><secondary-authors><author>Centers for Disease Control and Prevention</author></secondary-authors></contributors><titles><title><style face="italic" font="default" size="100%">Legionella</style><style face="normal" font="default" size="100%"> Home Page</style></title></titles><volume>2014</volume><number>January</number><dates><year>2013</year><pub-dates><date>February 5</date></pub-dates></dates><pub-location>Centers for Disease Control and Prevention</pub-location><publisher></publisher><urls><related-urls><url>;(CDC 2013). Sources that commonly become contaminated with legionella are cooling towers, hot water tanks, plumbing systems, fountains, and hot tubs. The bacteria can cause illness if contaminated water is inhaled or aspirated as mist, vapor, or droplets. The estimated fatality rate is 5-30% for those who get the disease ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2013</Year><RecNum>1</RecNum><DisplayText>(CDC 2013)</DisplayText><record><rec-number>1</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395349901">1</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author>CDC</author></authors><secondary-authors><author>Centers for Disease Control and Prevention</author></secondary-authors></contributors><titles><title><style face="italic" font="default" size="100%">Legionella</style><style face="normal" font="default" size="100%"> Home Page</style></title></titles><volume>2014</volume><number>January</number><dates><year>2013</year><pub-dates><date>February 5</date></pub-dates></dates><pub-location>Centers for Disease Control and Prevention</pub-location><publisher></publisher><urls><related-urls><url>;(CDC 2013). Especially at risk for the disease are adults aged 50 and older, males, smokers, diabetics, and people with chronic lung disease or any other sort of compromised immune system ADDIN EN.CITE <EndNote><Cite><Author>NIH</Author><Year>2014</Year><RecNum>14</RecNum><DisplayText>(NIH 2014)</DisplayText><record><rec-number>14</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395701167">14</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author>NIH</author></authors></contributors><titles><title>Legionnaires&apos; Disease</title></titles><volume>2014</volume><dates><year>2014</year><pub-dates><date>7 March 2014</date></pub-dates></dates><pub-location>US Library of Medicine</pub-location><publisher>MedlinePlus</publisher><urls><related-urls><url>;(NIH 2014). Comorbid conditions, the natural decline of the immune system with age, and decreased lung capacity are all contributors to the increased rate of Legionella and other bacterial infections that can lead to pneumonia in the elderly ADDIN EN.CITE <EndNote><Cite><Author>El-Solh</Author><Year>2001</Year><RecNum>16</RecNum><DisplayText>(El-Solh, Sikka et al. 2001)</DisplayText><record><rec-number>16</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395701569">16</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>El-Solh, Ali?A</author><author>Sikka, Pawan</author><author>Ramadan, Fadi</author><author>Davies, Joan</author></authors></contributors><titles><title>Etiology of Severe Pneumonia in the Very Elderly</title><secondary-title>American Journal of Respiratory and Critical Care Medicine</secondary-title></titles><periodical><full-title>American Journal of Respiratory and Critical Care Medicine</full-title></periodical><pages>645-651</pages><volume>163</volume><number>3</number><dates><year>2001</year><pub-dates><date>2001/03/01</date></pub-dates></dates><publisher>American Thoracic Society - AJRCCM</publisher><isbn>1073-449X</isbn><urls><related-urls><url>;(El-Solh, Sikka et al. 2001). Rates in younger people are much lower, ranging from one half to one tenth of the incident rate depending on the age bracket, and will almost always involve a comorbid condition or weakened immune system ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2011</Year><RecNum>13</RecNum><DisplayText>(CDC 2011)</DisplayText><record><rec-number>13</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395700999">13</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>CDC</author></authors></contributors><titles><title>Legionellosis-United States 2000-2009</title><secondary-title>Morbidity and Mortality Weekly Report</secondary-title></titles><pages>1083-1086</pages><volume>32</volume><num-vols>60</num-vols><dates><year>2011</year></dates><publisher>U.S. Department of Health and Human Services</publisher><urls></urls></record></Cite></EndNote>(CDC 2011).Incidence of legionnaire’s diseaseSince the discovery of this bacteria and its infection 37 years ago, the overall burden of reported legionellosis has risen over time in the United States, but remains relatively low overall ADDIN EN.CITE <EndNote><Cite><Author>Neil</Author><Year>2008</Year><RecNum>12</RecNum><DisplayText>(Neil and Berkelman 2008)</DisplayText><record><rec-number>12</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395700615">12</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Neil, Karen</author><author>Berkelman, Ruth</author></authors></contributors><titles><title>Increasing Incidence of Legionellosis in the United States, 1990–2005: Changing Epidemiologic Trends</title><secondary-title>Clinical Infectious Diseases</secondary-title></titles><periodical><full-title>Clinical Infectious Diseases</full-title></periodical><pages>591-599</pages><volume>47</volume><number>5</number><dates><year>2008</year><pub-dates><date>September 1, 2008</date></pub-dates></dates><urls><related-urls><url>;(Neil and Berkelman 2008)(Figure 1). A 2011 Morbidity and Mortality Weekly Report article stated that in the US, the incidence rate of reported legionellosis cases had increased from 0.4 cases per 100,000 people in 2000 to almost 1.1 cases per 100,000 people in 2009, which is a more than double increase in reported cases in less than a decade ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2011</Year><RecNum>13</RecNum><DisplayText>(CDC 2011)</DisplayText><record><rec-number>13</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395700999">13</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>CDC</author></authors></contributors><titles><title>Legionellosis-United States 2000-2009</title><secondary-title>Morbidity and Mortality Weekly Report</secondary-title></titles><pages>1083-1086</pages><volume>32</volume><num-vols>60</num-vols><dates><year>2011</year></dates><publisher>U.S. Department of Health and Human Services</publisher><urls></urls></record></Cite></EndNote>(CDC 2011). As a comparison, the rates of reported legionellosis in Allegheny County increased from around 4.5 cases per 100,000 people in 2003 to 7.6 cases per 100,000 people in 2013, meaning that in 2009, the Allegheny County rate of reported legionellosis was five times greater than the US rate (Figure 2).Figure SEQ Figure \* ARABIC 1. Annual number and age-adjusted incidence rate per 100,000 population of reported legionella cases in US, reported through the Centers for Disease Control for 1990-2005Figure SEQ Figure \* ARABIC 2. Annual reported rates of legionellosis, US and Allegheny CountyFor the elderly, pneumonia is the leading cause of death from an infectious disease and the fourth leading cause of death overall. While the aggregate rate of legionellosis was 0.75 cases per 100,000 people in the US from 2000-2009, the age-specific rate for the elderly in the US was 2.3 per 100,000 elderly people (age 60 and older) in that same period ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2011</Year><RecNum>13</RecNum><DisplayText>(CDC 2011)</DisplayText><record><rec-number>13</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395700999">13</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>CDC</author></authors></contributors><titles><title>Legionellosis-United States 2000-2009</title><secondary-title>Morbidity and Mortality Weekly Report</secondary-title></titles><pages>1083-1086</pages><volume>32</volume><num-vols>60</num-vols><dates><year>2011</year></dates><publisher>U.S. Department of Health and Human Services</publisher><urls></urls></record></Cite></EndNote>(CDC 2011). Figure 3 shows a comparison between the US and Allegheny County age-specific rates for older age groups. As can be seen, the Allegheny County rates are far above the US rates for all age groups, but in both populations, the rates continuously rise with age.Figure SEQ Figure \* ARABIC 3. Average age-specific rates of reported legionellosis cases in US (2000-2009) and Allegheny County (2003-2013); Note: different age classes are used for US and Allegheny County RatesIn a 2001 review of severe pneumonia in the elderly, El Sohl (et. al.) estimated that legionellosis account for 9% of community acquired cases, which would make it the second leading cause of pneumonia (behind Streptococcus pneumonia) among that population ADDIN EN.CITE <EndNote><Cite><Author>El-Solh</Author><Year>2001</Year><RecNum>16</RecNum><DisplayText>(El-Solh, Sikka et al. 2001)</DisplayText><record><rec-number>16</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395701569">16</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>El-Solh, Ali?A</author><author>Sikka, Pawan</author><author>Ramadan, Fadi</author><author>Davies, Joan</author></authors></contributors><titles><title>Etiology of Severe Pneumonia in the Very Elderly</title><secondary-title>American Journal of Respiratory and Critical Care Medicine</secondary-title></titles><periodical><full-title>American Journal of Respiratory and Critical Care Medicine</full-title></periodical><pages>645-651</pages><volume>163</volume><number>3</number><dates><year>2001</year><pub-dates><date>2001/03/01</date></pub-dates></dates><publisher>American Thoracic Society - AJRCCM</publisher><isbn>1073-449X</isbn><urls><related-urls><url>;(El-Solh, Sikka et al. 2001). In 2005, Seenivasan (et. al.) evaluated several outbreaks of Legionella illnesses specifically in long-term care facilities and found a few common underlying illnesses, such as stroke and chronic lung disease. They also found that overall legionellosis is very often under-diagnosed in these facilities due to insufficient specificity of the different diagnostic tests used and low use of testing. They suggested future case-control studies be done in these facilities ADDIN EN.CITE <EndNote><Cite><Author>Seenivasan</Author><Year>2005</Year><RecNum>2</RecNum><DisplayText>(Seenivasan, Yu et al. 2005)</DisplayText><record><rec-number>2</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395351153">2</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Seenivasan, Meena H.</author><author>Yu, Victor L.</author><author>Muder, Robert R.</author></authors></contributors><titles><title>Legionnaires&apos; Disease in Long-Term Care Facilities: Overview and Proposed Solutions</title><secondary-title>Journal of the American Geriatrics Society</secondary-title></titles><periodical><full-title>Journal of the American Geriatrics Society</full-title></periodical><pages>875-880</pages><volume>53</volume><number>5</number><keywords><keyword>Legionella</keyword><keyword>Legionnaires</keyword><keyword>disease</keyword><keyword>long-term care</keyword><keyword>pneumonia</keyword></keywords><dates><year>2005</year></dates><publisher>Blackwell Science Inc</publisher><isbn>1532-5415</isbn><urls><related-urls><url>;(Seenivasan, Yu et al. 2005). Two other publications more recently have highlighted outbreaks occurring in skilled nursing homes and senior high rises as well. For the purposes of this paper, senior high rises (SHR) will refer to privately or publically owned apartment buildings that are designated for senior citizens only. The first report by Skaza (et. al.) summarized a Slovenian skilled nursing home outbreak that was identified as the cause of at least 10 cases of legionellosis of various serotypes. Illnesses were most probably caused by a disturbance in the water flow of the building in 2010 PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5Ucm9wIFNrYXphPC9BdXRob3I+PFllYXI+MjAxMjwvWWVh

cj48UmVjTnVtPjE4PC9SZWNOdW0+PERpc3BsYXlUZXh0PihUcm9wIFNrYXphLCBCZXNrb3ZuaWsg

ZXQgYWwuIDIwMTIpPC9EaXNwbGF5VGV4dD48cmVjb3JkPjxyZWMtbnVtYmVyPjE4PC9yZWMtbnVt

YmVyPjxmb3JlaWduLWtleXM+PGtleSBhcHA9IkVOIiBkYi1pZD0iMnJ4MjBydHhoMHdlcjllenR4

aHAyZXBnOXNmd3I1dHRzZWZwIiB0aW1lc3RhbXA9IjEzOTU3MDMxNjEiPjE4PC9rZXk+PC9mb3Jl

aWduLWtleXM+PHJlZi10eXBlIG5hbWU9IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBlPjxj

b250cmlidXRvcnM+PGF1dGhvcnM+PGF1dGhvcj5Ucm9wIFNrYXphLCBBLjwvYXV0aG9yPjxhdXRo

b3I+QmVza292bmlrLCBMLjwvYXV0aG9yPjxhdXRob3I+U3Rvcm1hbiwgQS48L2F1dGhvcj48YXV0

aG9yPktlc2UsIEQuPC9hdXRob3I+PGF1dGhvcj5VcnNpYywgUy48L2F1dGhvcj48L2F1dGhvcnM+

PC9jb250cmlidXRvcnM+PGF1dGgtYWRkcmVzcz5JbnN0aXR1dGUgb2YgUHVibGljIEhlYWx0aCBD

ZWxqZSwgQ2VsamUsIFNsb3ZlbmlhLiBhbGVua2Euc2themFAenp2LWNlLnNpPC9hdXRoLWFkZHJl

c3M+PHRpdGxlcz48dGl0bGU+RXBpZGVtaW9sb2dpY2FsIGludmVzdGlnYXRpb24gb2YgYSBsZWdp

b25lbGxvc2lzIG91dGJyZWFrIGluIGEgU2xvdmVuaWFuIG51cnNpbmcgaG9tZSwgQXVndXN0IDIw

MTA8L3RpdGxlPjxzZWNvbmRhcnktdGl0bGU+U2NhbmQgSiBJbmZlY3QgRGlzPC9zZWNvbmRhcnkt

dGl0bGU+PGFsdC10aXRsZT5TY2FuZGluYXZpYW4gam91cm5hbCBvZiBpbmZlY3Rpb3VzIGRpc2Vh

c2VzPC9hbHQtdGl0bGU+PC90aXRsZXM+PHBlcmlvZGljYWw+PGZ1bGwtdGl0bGU+U2NhbmQgSiBJ

bmZlY3QgRGlzPC9mdWxsLXRpdGxlPjxhYmJyLTE+U2NhbmRpbmF2aWFuIGpvdXJuYWwgb2YgaW5m

ZWN0aW91cyBkaXNlYXNlczwvYWJici0xPjwvcGVyaW9kaWNhbD48YWx0LXBlcmlvZGljYWw+PGZ1

bGwtdGl0bGU+U2NhbmQgSiBJbmZlY3QgRGlzPC9mdWxsLXRpdGxlPjxhYmJyLTE+U2NhbmRpbmF2

aWFuIGpvdXJuYWwgb2YgaW5mZWN0aW91cyBkaXNlYXNlczwvYWJici0xPjwvYWx0LXBlcmlvZGlj

YWw+PHBhZ2VzPjI2My05PC9wYWdlcz48dm9sdW1lPjQ0PC92b2x1bWU+PG51bWJlcj40PC9udW1i

ZXI+PGVkaXRpb24+MjAxMi8wMi8yMjwvZWRpdGlvbj48a2V5d29yZHM+PGtleXdvcmQ+QWR1bHQ8

L2tleXdvcmQ+PGtleXdvcmQ+RGlzZWFzZSBPdXRicmVha3MvKnN0YXRpc3RpY3MgJmFtcDsgbnVt

ZXJpY2FsIGRhdGE8L2tleXdvcmQ+PGtleXdvcmQ+RHJpbmtpbmcgV2F0ZXIvbWljcm9iaW9sb2d5

PC9rZXl3b3JkPjxrZXl3b3JkPkVudmlyb25tZW50YWwgTWljcm9iaW9sb2d5PC9rZXl3b3JkPjxr

ZXl3b3JkPkZlbWFsZTwva2V5d29yZD48a2V5d29yZD5IdW1hbnM8L2tleXdvcmQ+PGtleXdvcmQ+

TGVnaW9uZWxsYSBwbmV1bW9waGlsYS9nZW5ldGljcy8qaXNvbGF0aW9uICZhbXA7IHB1cmlmaWNh

dGlvbjwva2V5d29yZD48a2V5d29yZD5MZWdpb25uYWlyZXMmYXBvczsgRGlzZWFzZS8qZXBpZGVt

aW9sb2d5L21pY3JvYmlvbG9neTwva2V5d29yZD48a2V5d29yZD5NYWxlPC9rZXl3b3JkPjxrZXl3

b3JkPk1pZGRsZSBBZ2VkPC9rZXl3b3JkPjxrZXl3b3JkPk51cnNpbmcgSG9tZXMvKnN0YXRpc3Rp

Y3MgJmFtcDsgbnVtZXJpY2FsIGRhdGE8L2tleXdvcmQ+PGtleXdvcmQ+UXVlc3Rpb25uYWlyZXM8

L2tleXdvcmQ+PGtleXdvcmQ+U2xvdmVuaWEvZXBpZGVtaW9sb2d5PC9rZXl3b3JkPjwva2V5d29y

ZHM+PGRhdGVzPjx5ZWFyPjIwMTI8L3llYXI+PHB1Yi1kYXRlcz48ZGF0ZT5BcHI8L2RhdGU+PC9w

dWItZGF0ZXM+PC9kYXRlcz48aXNibj4wMDM2LTU1NDg8L2lzYm4+PGFjY2Vzc2lvbi1udW0+MjIz

Mzk1NDE8L2FjY2Vzc2lvbi1udW0+PHVybHM+PC91cmxzPjxlbGVjdHJvbmljLXJlc291cmNlLW51

bT4xMC4zMTA5LzAwMzY1NTQ4LjIwMTEuNjM1MzEzPC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48

cmVtb3RlLWRhdGFiYXNlLXByb3ZpZGVyPk5MTTwvcmVtb3RlLWRhdGFiYXNlLXByb3ZpZGVyPjxs

YW5ndWFnZT5lbmc8L2xhbmd1YWdlPjwvcmVjb3JkPjwvQ2l0ZT48L0VuZE5vdGU+AG==

ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5Ucm9wIFNrYXphPC9BdXRob3I+PFllYXI+MjAxMjwvWWVh

cj48UmVjTnVtPjE4PC9SZWNOdW0+PERpc3BsYXlUZXh0PihUcm9wIFNrYXphLCBCZXNrb3ZuaWsg

ZXQgYWwuIDIwMTIpPC9EaXNwbGF5VGV4dD48cmVjb3JkPjxyZWMtbnVtYmVyPjE4PC9yZWMtbnVt

YmVyPjxmb3JlaWduLWtleXM+PGtleSBhcHA9IkVOIiBkYi1pZD0iMnJ4MjBydHhoMHdlcjllenR4

aHAyZXBnOXNmd3I1dHRzZWZwIiB0aW1lc3RhbXA9IjEzOTU3MDMxNjEiPjE4PC9rZXk+PC9mb3Jl

aWduLWtleXM+PHJlZi10eXBlIG5hbWU9IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBlPjxj

b250cmlidXRvcnM+PGF1dGhvcnM+PGF1dGhvcj5Ucm9wIFNrYXphLCBBLjwvYXV0aG9yPjxhdXRo

b3I+QmVza292bmlrLCBMLjwvYXV0aG9yPjxhdXRob3I+U3Rvcm1hbiwgQS48L2F1dGhvcj48YXV0

aG9yPktlc2UsIEQuPC9hdXRob3I+PGF1dGhvcj5VcnNpYywgUy48L2F1dGhvcj48L2F1dGhvcnM+

PC9jb250cmlidXRvcnM+PGF1dGgtYWRkcmVzcz5JbnN0aXR1dGUgb2YgUHVibGljIEhlYWx0aCBD

ZWxqZSwgQ2VsamUsIFNsb3ZlbmlhLiBhbGVua2Euc2themFAenp2LWNlLnNpPC9hdXRoLWFkZHJl

c3M+PHRpdGxlcz48dGl0bGU+RXBpZGVtaW9sb2dpY2FsIGludmVzdGlnYXRpb24gb2YgYSBsZWdp

b25lbGxvc2lzIG91dGJyZWFrIGluIGEgU2xvdmVuaWFuIG51cnNpbmcgaG9tZSwgQXVndXN0IDIw

MTA8L3RpdGxlPjxzZWNvbmRhcnktdGl0bGU+U2NhbmQgSiBJbmZlY3QgRGlzPC9zZWNvbmRhcnkt

dGl0bGU+PGFsdC10aXRsZT5TY2FuZGluYXZpYW4gam91cm5hbCBvZiBpbmZlY3Rpb3VzIGRpc2Vh

c2VzPC9hbHQtdGl0bGU+PC90aXRsZXM+PHBlcmlvZGljYWw+PGZ1bGwtdGl0bGU+U2NhbmQgSiBJ

bmZlY3QgRGlzPC9mdWxsLXRpdGxlPjxhYmJyLTE+U2NhbmRpbmF2aWFuIGpvdXJuYWwgb2YgaW5m

ZWN0aW91cyBkaXNlYXNlczwvYWJici0xPjwvcGVyaW9kaWNhbD48YWx0LXBlcmlvZGljYWw+PGZ1

bGwtdGl0bGU+U2NhbmQgSiBJbmZlY3QgRGlzPC9mdWxsLXRpdGxlPjxhYmJyLTE+U2NhbmRpbmF2

aWFuIGpvdXJuYWwgb2YgaW5mZWN0aW91cyBkaXNlYXNlczwvYWJici0xPjwvYWx0LXBlcmlvZGlj

YWw+PHBhZ2VzPjI2My05PC9wYWdlcz48dm9sdW1lPjQ0PC92b2x1bWU+PG51bWJlcj40PC9udW1i

ZXI+PGVkaXRpb24+MjAxMi8wMi8yMjwvZWRpdGlvbj48a2V5d29yZHM+PGtleXdvcmQ+QWR1bHQ8

L2tleXdvcmQ+PGtleXdvcmQ+RGlzZWFzZSBPdXRicmVha3MvKnN0YXRpc3RpY3MgJmFtcDsgbnVt

ZXJpY2FsIGRhdGE8L2tleXdvcmQ+PGtleXdvcmQ+RHJpbmtpbmcgV2F0ZXIvbWljcm9iaW9sb2d5

PC9rZXl3b3JkPjxrZXl3b3JkPkVudmlyb25tZW50YWwgTWljcm9iaW9sb2d5PC9rZXl3b3JkPjxr

ZXl3b3JkPkZlbWFsZTwva2V5d29yZD48a2V5d29yZD5IdW1hbnM8L2tleXdvcmQ+PGtleXdvcmQ+

TGVnaW9uZWxsYSBwbmV1bW9waGlsYS9nZW5ldGljcy8qaXNvbGF0aW9uICZhbXA7IHB1cmlmaWNh

dGlvbjwva2V5d29yZD48a2V5d29yZD5MZWdpb25uYWlyZXMmYXBvczsgRGlzZWFzZS8qZXBpZGVt

aW9sb2d5L21pY3JvYmlvbG9neTwva2V5d29yZD48a2V5d29yZD5NYWxlPC9rZXl3b3JkPjxrZXl3

b3JkPk1pZGRsZSBBZ2VkPC9rZXl3b3JkPjxrZXl3b3JkPk51cnNpbmcgSG9tZXMvKnN0YXRpc3Rp

Y3MgJmFtcDsgbnVtZXJpY2FsIGRhdGE8L2tleXdvcmQ+PGtleXdvcmQ+UXVlc3Rpb25uYWlyZXM8

L2tleXdvcmQ+PGtleXdvcmQ+U2xvdmVuaWEvZXBpZGVtaW9sb2d5PC9rZXl3b3JkPjwva2V5d29y

ZHM+PGRhdGVzPjx5ZWFyPjIwMTI8L3llYXI+PHB1Yi1kYXRlcz48ZGF0ZT5BcHI8L2RhdGU+PC9w

dWItZGF0ZXM+PC9kYXRlcz48aXNibj4wMDM2LTU1NDg8L2lzYm4+PGFjY2Vzc2lvbi1udW0+MjIz

Mzk1NDE8L2FjY2Vzc2lvbi1udW0+PHVybHM+PC91cmxzPjxlbGVjdHJvbmljLXJlc291cmNlLW51

bT4xMC4zMTA5LzAwMzY1NTQ4LjIwMTEuNjM1MzEzPC9lbGVjdHJvbmljLXJlc291cmNlLW51bT48

cmVtb3RlLWRhdGFiYXNlLXByb3ZpZGVyPk5MTTwvcmVtb3RlLWRhdGFiYXNlLXByb3ZpZGVyPjxs

YW5ndWFnZT5lbmc8L2xhbmd1YWdlPjwvcmVjb3JkPjwvQ2l0ZT48L0VuZE5vdGU+AG==

ADDIN EN.CITE.DATA (Trop Skaza, Beskovnik et al. 2012). In the second report from 2013, Silk (et. al.) described a Baltimore SHR outbreak of Legionella causing illness in 8 residents and 2 visitors in which remediation techniques were outlined and recommended for follow-up and prevention for future such outbreaks ADDIN EN.CITE <EndNote><Cite><Author>Silk</Author><Year>2013</Year><RecNum>3</RecNum><DisplayText>(Silk, Foltz et al. 2013)</DisplayText><record><rec-number>3</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395352000">3</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Silk, B. J.</author><author>Foltz, J. L.</author><author>Ngamsnga, K.</author><author>Brown, E.</author><author>Munoz, M. G.</author><author>Hampton, L.</author><author>Jacobs-Slifka, K.</author><author>Kozak, N. A.</author><author>Underwood, J. M.</author><author>Krick, J.</author><author>Travis, T.</author><author>Farrow, O.</author><author>Fields, B. S.</author><author>Blythe, D.</author><author>Hicks, L. A.</author></authors></contributors><titles><title>Legionnaires&apos; disease case-finding algorithm, attack rates, and risk factors during a residential outbreak among older adults: an environmental and cohort study</title><secondary-title>BMC Infect Dis</secondary-title><alt-title>BMC infectious diseases</alt-title></titles><periodical><full-title>BMC Infect Dis</full-title><abbr-1>BMC infectious diseases</abbr-1></periodical><alt-periodical><full-title>BMC Infect Dis</full-title><abbr-1>BMC infectious diseases</abbr-1></alt-periodical><pages>291</pages><volume>13</volume><number>1</number><edition>2013/06/29</edition><dates><year>2013</year><pub-dates><date>Jun 27</date></pub-dates></dates><isbn>1471-2334</isbn><accession-num>23806063</accession-num><urls></urls><custom2>Pmc3700825</custom2><electronic-resource-num>10.1186/1471-2334-13-291</electronic-resource-num><remote-database-provider>NLM</remote-database-provider><language>Eng</language></record></Cite></EndNote>(Silk, Foltz et al. 2013).In Allegheny County, the elderly (people age 65 and older) comprise approximately 17% of the overall population, which is higher than the US elderly proportion of 13% ADDIN EN.CITE <EndNote><Cite><Author>Howden</Author><Year>2011</Year><RecNum>26</RecNum><DisplayText>(Howden and Meyer 2011, Kokenda 2013)</DisplayText><record><rec-number>26</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1398200004">26</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Lindsay Howden</author><author>Julia Meyer</author></authors></contributors><titles><title>Age and Sex Composition: 2010</title><secondary-title>2010 Census Briefs</secondary-title></titles><dates><year>2011</year></dates><publisher>US Department of Commerce-US Census Bureau</publisher><urls></urls></record></Cite><Cite><Author>Kokenda</Author><Year>2013</Year><RecNum>27</RecNum><record><rec-number>27</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1398200096">27</key></foreign-keys><ref-type name="Aggregated Database">55</ref-type><contributors><authors><author>John Kokenda</author></authors></contributors><titles><title>Allegheny County Census Data</title></titles><dates><year>2013</year></dates><pub-location>Allegheny County Health Department</pub-location><urls></urls></record></Cite></EndNote>(Howden and Meyer 2011, Kokenda 2013). It was reported in 2010 that approximately 5% of elderly people in the county are living in assisted living or long-term care facilities ADDIN EN.CITE <EndNote><Cite><Author>Musa</Author><Year>2003</Year><RecNum>25</RecNum><DisplayText>(Musa, Seiler et al. 2003)</DisplayText><record><rec-number>25</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1398197772">25</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Donald Musa</author><author>Jason Seiler</author><author>Paul Flora</author><author>Christopher Briem</author><author>Lynn Martire</author><author>Richard Schulz</author></authors></contributors><titles><title>The State of Aging and Health in Pittsburgh and Allegheny County</title></titles><dates><year>2003</year></dates><publisher>University of Pittsburgh Graduate School of Public Health</publisher><urls></urls></record></Cite></EndNote>(Musa, Seiler et al. 2003). This rate is also higher than the US proportion of elderly in long-term care facilities, which is estimated to be 4% ADDIN EN.CITE <EndNote><Cite><Author>Harris-Kojetin</Author><Year>2013</Year><RecNum>28</RecNum><DisplayText>(Harris-Kojetin, Sengupta et al. 2013)</DisplayText><record><rec-number>28</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1398200615">28</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>L Harris-Kojetin</author><author>M Sengupta</author><author>E Park-Lee</author><author>R Valverde</author></authors></contributors><titles><title>Long-term care services in the United State: 2013 overview</title></titles><dates><year>2013</year></dates><pub-location>Hyattsville, MD</pub-location><publisher>Center for Health Statistics</publisher><urls></urls></record></Cite></EndNote>(Harris-Kojetin, Sengupta et al. 2013). This means that there are not only a greater proportion of elderly in Allegheny County, but a greater proportion of them living in long-term care facilities, which is a setting believed to be particularly susceptible for developing cases of legionellosis. This population, therefore, is of particular importance when studying and determining effective methods of testing, remediation, and prevention in Allegheny County.Surveillance of legionnaire’s diseaseSince the first outbreak and discovery of Legionella in 1976 at a hotel convention in Philadelphia, there have been major advances in scientific knowledge about the disease, including the creation and improvements of diagnostic tests. What was once considered an exotic disease has now been discovered to be a common form of severe pneumonia. There is still a serious lack of information though due to undiagnosed cases and incomplete surveillance of the disease. Although there are currently 48 species and 70 serotypes of the bacteria, the urine antigen test commonly used for diagnosis is intended to test for only one serotype of one species. And while the specificity of the diagnostic test allows it to identify some of the other serotypes, it is estimated to miss 30-40% of cases due to other serotypes. And, of the many pneumonia cases that present in hospitals and other clinical settings each year, not all will be tested to determine the specific bacteria causing the illness. Therefore, only a small proportion of cases are actually reported to the national and state surveillance systems each year ADDIN EN.CITE <EndNote><Cite><Author>Fields</Author><Year>2002</Year><RecNum>15</RecNum><DisplayText>(Fields, Benson et al. 2002)</DisplayText><record><rec-number>15</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395701493">15</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Fields, B. S.</author><author>Benson, R. F.</author><author>Besser, R. E.</author></authors></contributors><auth-address>Respiratory Diseases Branch, Division of Bacterial and Mycotic Diseases, National Center for Infectious Disease, Centers for Disease Control and Prevention, Atlanta, Georgia 30333, USA. bfeilds@</auth-address><titles><title>Legionella and Legionnaires&apos; disease: 25 years of investigation</title><secondary-title>Clin Microbiol Rev</secondary-title><alt-title>Clinical microbiology reviews</alt-title></titles><periodical><full-title>Clin Microbiol Rev</full-title><abbr-1>Clinical microbiology reviews</abbr-1></periodical><alt-periodical><full-title>Clin Microbiol Rev</full-title><abbr-1>Clinical microbiology reviews</abbr-1></alt-periodical><pages>506-26</pages><volume>15</volume><number>3</number><edition>2002/07/05</edition><keywords><keyword>Animals</keyword><keyword>Humans</keyword><keyword>*Legionella/classification/pathogenicity/physiology</keyword><keyword>*Legionnaires&apos; Disease/diagnosis/epidemiology/microbiology/physiopathology</keyword><keyword>Research</keyword><keyword>Virulence</keyword></keywords><dates><year>2002</year><pub-dates><date>Jul</date></pub-dates></dates><isbn>0893-8512 (Print)&#xD;0893-8512</isbn><accession-num>12097254</accession-num><urls></urls><custom2>Pmc118082</custom2><remote-database-provider>NLM</remote-database-provider><language>eng</language></record></Cite></EndNote>(Fields, Benson et al. 2002). Furthermore, the cases that are reported may not be representative of the actual cases occurring since commonly identified cases are often immune-compromised and more likely to be diagnosed or cases living in large buildings where legionella would be more to be tested for and identified ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2013</Year><RecNum>1</RecNum><DisplayText>(CDC 2013)</DisplayText><record><rec-number>1</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395349901">1</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author>CDC</author></authors><secondary-authors><author>Centers for Disease Control and Prevention</author></secondary-authors></contributors><titles><title><style face="italic" font="default" size="100%">Legionella</style><style face="normal" font="default" size="100%"> Home Page</style></title></titles><volume>2014</volume><number>January</number><dates><year>2013</year><pub-dates><date>February 5</date></pub-dates></dates><pub-location>Centers for Disease Control and Prevention</pub-location><publisher></publisher><urls><related-urls><url>;(CDC 2013).Evaluating a Surveillance SystemIn 1988, the Centers for Disease Control and Prevention (CDC) first published their Guidelines for Evaluating a Surveillance System. This document quickly became the standard used to determine the efficiency and effectiveness of public health surveillance systems. This document was updated in 2001 to adapt to the growing used of electronic documenting in both the clinical and public health fields. When evaluating a surveillance system by the new CDC guidelines, there are several key system attributes that should be review and analyzed: simplicity, flexibility, data quality, acceptability, sensitivity, predictive value positive, representativeness, timeliness, and stability ADDIN EN.CITE <EndNote><Cite><Author>German</Author><Year>2001</Year><RecNum>21</RecNum><DisplayText>(German, Lee et al. 2001)</DisplayText><record><rec-number>21</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1396474810">21</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Robert R. German</author><author>Lisa M. Lee</author><author>John M. Horan</author><author>Robert L. Milstein</author><author>Carol A. Pertowski</author><author>Michael N. Waller</author></authors></contributors><titles><title>Updated Guidelines for Evaluating Public Health Surveillance Systems</title><secondary-title>Morbitity and Mortality Weekly Report</secondary-title></titles><pages>1-35</pages><volume>50</volume><number>RR13</number><dates><year>2001</year></dates><publisher>Department of Health and Human Services</publisher><urls></urls></record></Cite></EndNote>(German, Lee et al. 2001). Following are the paraphrased definitions of each attribute from the updated document:Simplicity: the structure and ease of operation of a system; surveillance systems should be as simple as possible while still meeting their objectiveFlexibility: ability to adapt to changing information or conditions with little additional time, funds, and personnelData Quality: the completeness and validity of the data recorded in the public health surveillance systemAcceptability: the willingness or persons and organizations to participate in the surveillance systemSensitivity:The proportion of cases of a disease detected by the surveillance systemThe ability to detect outbreaks, including ability to monitor changes in the number of cases over timePredictive Positive Value: proportion of reported cases under surveillance that actually have the health-related event or diseaseRepresentativeness: accurately describing the occurrence of disease over time and its distribution in the populationTimeliness: the speed between steps in a public health surveillance systemStability: the ability of a public health surveillance system to collect, manage, and provide data properly without failure and its ability to operate when neededThese attributes are good to keep in mind not only when evaluating a surveillance system, but also when using a surveillance system in general. They were very useful during this legionellosis project, and ended up being evaluated in the second stage of the project when data quality and a few other attributes turned out to be lacking for the Allegheny County legionellosis surveillance system.guidelines for legionnaire’s diseaseWhen it comes to detecting and removing Legionella in a water source or system, there are a few guidelines available from government organizations, but they are often limited in their scope ADDIN EN.CITE <EndNote><Cite><Author></Author><Year>2012</Year><RecNum>19</RecNum><DisplayText>( 2012)</DisplayText><record><rec-number>19</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395703605">19</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author></author></authors></contributors><titles><title>Guidelines</title></titles><volume>2014</volume><dates><year>2012</year></dates><publisher></publisher><urls><related-urls><url>;( 2012). For the US, the Centers for Disease Control and Prevention (CDC) created guidelines for outbreaks in 2005, which includes proper collection and preparation of water specimens and the techniques for creating the necessary media and plating cultures ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2005</Year><RecNum>8</RecNum><DisplayText>(CDC 2005)</DisplayText><record><rec-number>8</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395353902">8</key></foreign-keys><ref-type name="Government Document">46</ref-type><contributors><authors><author>CDC</author></authors><secondary-authors><author>U.S. Department of Health and Human Services</author></secondary-authors></contributors><titles><title><style face="normal" font="default" size="100%">Procedures for the Recovery of </style><style face="italic" font="default" size="100%">Legionella</style><style face="normal" font="default" size="100%"> from the Environment</style></title></titles><dates><year>2005</year></dates><pub-location>Atlanta, Georgia</pub-location><publisher>Centers for Disease Control and Prevention</publisher><urls></urls></record></Cite></EndNote>(CDC 2005). The CDC does not, however, have guidelines on how to test or maintain an overall water system before a case occurs or when a non-outbreak case occurs. To fill this gap in information, several health departments have created local guidelines for testing, maintaining, and remediating Legionella from water systems. The first to do this was Allegheny County Health Department (ACHD) in 1997 in response to the large number and burden of cases in hospitals and other medical settings (Allegheny County acquires a disproportionately high number of the cases occurring in Pennsylvania). Their guidelines are aimed specifically at health-care facilities and include procedures for clinical diagnosis, environmental sampling (although different from the CDC’s suggestions), maintenance techniques, and several options for disinfection of a water supply, which included copper-silver ionization systems, heat and flush method, chlorination, flash heating systems, and ultraviolet irradiation ADDIN EN.CITE <EndNote><Cite><Author>ACHD</Author><Year>1997</Year><RecNum>9</RecNum><DisplayText>(ACHD 1997)</DisplayText><record><rec-number>9</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395699536">9</key></foreign-keys><ref-type name="Pamphlet">24</ref-type><contributors><authors><author>ACHD</author></authors></contributors><titles><title>Approaches to Prevention and Control of Legionella Infection in Allegheny County Health Care Facilities</title></titles><dates><year>1997</year></dates><pub-location>Pittsburgh</pub-location><publisher>Allegheny County Health Department</publisher><urls></urls></record></Cite></EndNote>(ACHD 1997).Two other health departments, Maryland and New York state, issued reports or guidelines on Legionella and Legionnaire’s Disease. In 1999 (updated in 2005), the New York State Health Department sent a report to all state hospitals, which outlined diagnosis, surveillance, environmental testing, and control measures ADDIN EN.CITE <EndNote><Cite><Author>Novello</Author><Year>2005</Year><RecNum>5</RecNum><DisplayText>(Novello 2005)</DisplayText><record><rec-number>5</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395353391">5</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Antonia Novello</author></authors></contributors><titles><title>Hospital-associated Legionellosis</title></titles><dates><year>2005</year></dates><pub-location>Albany, NY</pub-location><publisher>State of New York Department of Health</publisher><urls></urls></record></Cite></EndNote>(Novello 2005). In 2000, the Maryland Department of Health and Mental Hygiene issued a special report, based primarily on those of ASHRAE and the Allegheny County Health Department, that gave a summary of how Maryland health-care facilities and other buildings with vulnerable populations should handle risk assessment, surveillance, testing, and remediation. No blanket guidelines were given, but instead recommended options were outlined for the decision of each facility. The remediation techniques supported by both of these reports were heat and flush, chlorination, and copper-silver ionization, while Maryland additionally listed monochloramine and ultraviolet treatments ADDIN EN.CITE <EndNote><Cite><Author>MDHMH</Author><Year>2000</Year><RecNum>4</RecNum><DisplayText>(MDHMH 2000)</DisplayText><record><rec-number>4</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395352233">4</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>MDHMH</author></authors></contributors><titles><title>Report of the Maryland Scientific Working Group to&#xD;Study Legionella in Water Systems in Healthcare Institutions</title></titles><dates><year>2000</year></dates><pub-location>Baltimore, Maryland</pub-location><publisher>Maryland Department of Health and Mental Hygiene</publisher><urls></urls></record></Cite></EndNote>(MDHMH 2000).The American Society of Heating, Refrigerating, and Air-Conditioning Engineers (ASHRAE) originally wrote one of the most comprehensive guideline documents in 2000. They are currently updating their guidelines to include the most recent information for the maintenance, detection and remediation of Legionella in water systems of large buildings. While still officially under review, the document outlines both the heat and flush method and chemical treatment using an EPA approved halogen, such as chlorine, for water in the circumstance of any established case(s), but also outlines the development of a Hazard Analysis and Critical Control Point (HACCP) plan, routine testing procedures, and proper maintenance of a given water system. Separate plans are outlined for hot-water tanks, cold-water tanks, cooling towers or condensers, whirlpool spas, fountains or water features, misters, atomizers, air washers, and humidifiers ADDIN EN.CITE <EndNote><Cite><Author>ASHRAE</Author><Year>2013</Year><RecNum>11</RecNum><DisplayText>(ASHRAE 2013)</DisplayText><record><rec-number>11</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395700355">11</key></foreign-keys><ref-type name="Standard">58</ref-type><contributors><authors><author>ASHRAE</author></authors></contributors><titles><title>Prevention of Legionellosis Assiciated with Building Water Systems</title></titles><volume>BSR/ASHRAE Standard 188P</volume><dates><year>2013</year></dates><pub-location>Atlanta, GA</pub-location><publisher>ASHRAE</publisher><urls></urls></record></Cite></EndNote>(ASHRAE 2013).Two other documents of note were produced. The World Health Organization (WHO) released the first in 2007 and the second was released by several health agencies of the South Australian government in 2008. The WHO document outlines the necessity for a “water safety plan”, similar to a HACCP, for water plumbing in buildings, cooling towers/condensers, health-care facilities, hotels and ships, and hot tubs or swimming pools ADDIN EN.CITE <EndNote><Cite><Author>WHO</Author><Year>2007</Year><RecNum>7</RecNum><DisplayText>(WHO 2007)</DisplayText><record><rec-number>7</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395353692">7</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>WHO</author></authors><secondary-authors><author>Jamie Bartram</author><author>Yves Chartier</author><author>John V Lee</author><author>Kathy Pond</author><author>Susanne Surman-Lee</author></secondary-authors></contributors><titles><title>Legionella and the Prevention of Legionellosis</title></titles><dates><year>2007</year></dates><pub-location>India</pub-location><publisher>World Health Organization</publisher><urls></urls></record></Cite></EndNote>(WHO 2007). They give a brief analysis of 14 different remediation techniques as well. The South Australian government document summarizes maintenance for usual water systems and also the more recently realized source of ice machines. The only remediation techniques allowed by their government are the heat and flush method and chlorination ADDIN EN.CITE <EndNote><Cite><Author>SAHealth</Author><Year>2008</Year><RecNum>10</RecNum><DisplayText>(SAHealth 2008)</DisplayText><record><rec-number>10</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395700009">10</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>SAHealth</author></authors></contributors><titles><title>Guidelines for the control of legionella in manufactured water systems in South Australia</title></titles><edition>Revised 2013</edition><dates><year>2008</year></dates><publisher>Government of South Australia, Health Protection Programs, Public Health Services, Public Health and Clinical Systems, and Department for Health and Ageing</publisher><urls></urls></record></Cite></EndNote>(SAHealth 2008).In light of the inconclusiveness and disagreement between organizations as to how to properly treat water, a few researchers are asking for more research to be done in the field of Legionella testing and remediation ADDIN EN.CITE <EndNote><Cite><Author>Stout</Author><Year>2013</Year><RecNum>23</RecNum><DisplayText>(Stout 2013)</DisplayText><record><rec-number>23</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1397082778">23</key></foreign-keys><ref-type name="Interview">64</ref-type><contributors><authors><author>Janet Stout</author></authors><secondary-authors><author>Lindsay Hamitlon</author></secondary-authors></contributors><titles><title>Planning Meeting for Legionella Paper</title><secondary-title>Allegheny County Health Department</secondary-title></titles><dates><year>2013</year></dates><urls></urls></record></Cite></EndNote>(Stout 2013). Therefore, in partnership with the Allegheny County Health Department, a project was started which would identify and analyze cases of legionellosis occurring in the elderly in both long term care facility and SHR settings as documented by the Pennsylvania National Electronic Disease Surveillance System (PA-NEDSS). Due to the large number of cases occurring in Pennsylvania and Allegheny County and in the elderly in particular, the objective of this project was to assess risk factors for legionellosis as well as the effectiveness of prevention and control measures taken in facilities by facilities caring for or housing the elderly. The goals of the initial study were to 1) find the odds of having certain risk factors when developing legionellosis, 2) find the relative risk of having a repeat case in a long-term care facility or SHR after environmental testing, and 3) find the relative risk of having a repeat case in a long-term care facility after a given remediation recommendation.When it became apparent that there was too much missing or misclassified data, the focus of the project shifted and its goals became to 1) assess the amount or proportion of missing data for all variables of interest, 2) analyze the legionellosis surveillance system, and 3) make recommendations based on conclusions of surveillance system. Risk Factors, Environmental Testing, And REmediationMethodsI obtained data from the Pennsylvania National Electronic Disease Surveillance System (PA-NEDSS), which is the database in which all notifiable illnesses are reported and investigated by health-care and public health workers respectively on both a local and state level. The data included all confirmed legionellosis cases from January 1, 2003 to December 31, 2013 that were reported in Allegheny County residents. Formatting and analysis was performed using SAS Enterprise Guide, SAS 9.3, and Microsoft Excel. Health-care exposure was designated as a selection of one or more of these nosocomial and risk factor variables: ‘definite nosocomial’, ‘probable nosocomial’, and/or ‘visited or stayed in health-care facility’.Definite nosocomial is defined by PA-NEDSS as a case living in any sort of health-care facility (not limited to hospital facilities, but only including those who were patients in a health-care facility) for the entirety of the incubation period, which is defined as 2-10 days prior to onset of symptoms. Probable nosocomial is defined as a case living or visiting any sort of health-care facility (inpatient or outpatient treatment) during incubation period. Visited or stayed in a health-care facility is defined as a case having any exposure to a health-care facility through visiting patient, working, or being treated as inpatient/outpatient. This risk factor variable is not mutually exclusive with either of the nosocomial variables above.To generate a list of those specifically exposed to long-term care facilities, cases were verified by reviewing investigation notes, ruling out misclassifications by comparing the case notes and risk factor variables selected, and cross-referencing the case addresses with the health department’s address list of skilled nursing and assisted living homes. Long-term care facilities were defined as any sort or skilled nursing home, assisted living center, or health rehabilitation center, and ‘long-term facility cases’ were defined as anyone exposed to their facility during their incubation period. For a full listing of risk factor variables included in the form, see Appendix.In order to produce the list of cases who were living in a SHR, a filter of 65 years old and up was used and addresses were cross-referenced with known SHR or independent living apartment buildings in Allegheny County according to “Senior Housing Net” website. ‘Senior high rise cases’ were defined as anyone exposed to their apartment during their incubation period. The investigation notes contained in PA-NEDSS were again used to verify cases.All long-term care facility and SHR legionellosis case investigations were reviewed to determine if any environmental (water and surface swab) testing was performed and, if so, what recommendations were made to the facility for possible remediation. resultsThe following table and figure outline the overall characteristics of all legionellosis cases. The variables have both the total number and percent, with the overall total of the sample listed at the top of the table.Table SEQ Table \* ARABIC 1. Summary characteristics for total Allegheny County reported legionellosis cases(Total N=835)Total (n)Percentage (%)CharacteristicsFemale32337.9Male53062.1Age 60+51360.1Age 65+41849.0Age 70+33739.5Age 80+17320.3Risk FactorsSenior High Rise Exposed182.1Health-care Facility Employee182.1Previous Overnight Travel12314.4Previous Dental Work202.3Health-Care Exposed13215.8Acute Care Exposed829.9Long-Term Care Exposed425.0Both70.8In the above table (Table 1), ‘health-care exposed’ is defined as a case that has lived, stayed, or been in a health-care setting during their determined incubation period and confirmed by case notes, and mostly consists of employees or patients of health-care facilities. This category is in contrast to the alternative classification of ‘community acquired’ cases.Figure SEQ Figure \* ARABIC 4. Annual age-specific rates of reported legionellosis in Allegheny County from 2003-2013Figure 4 summarizes the findings of age distribution of legionellosis cases. These rates are age-specific and therefore not adjusted.The goal of calculating the odds and relative risk ratios, however, was never completed due to the fact that a large portion of testing and recommendation data was missing. Upon this realization that too much data were missing, impartial, or misclassified, the scope of conclusions to be drawn was changed.Analysis of Allegheny COunty Legionellosis surveillancemethodsUpon the realization that too much data was missing, it was decided that a qualitative assessment of the legionellosis surveillance system would be performed. When evaluating a surveillance system by the CDC guidelines, there are several key system attributes that should be review and analyzed: simplicity, flexibility, data quality, acceptability, sensitivity, predictive value positive, representativeness, timeliness, and stability ADDIN EN.CITE <EndNote><Cite><Author>German</Author><Year>2001</Year><RecNum>21</RecNum><DisplayText>(German, Lee et al. 2001)</DisplayText><record><rec-number>21</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1396474810">21</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Robert R. German</author><author>Lisa M. Lee</author><author>John M. Horan</author><author>Robert L. Milstein</author><author>Carol A. Pertowski</author><author>Michael N. Waller</author></authors></contributors><titles><title>Updated Guidelines for Evaluating Public Health Surveillance Systems</title><secondary-title>Morbitity and Mortality Weekly Report</secondary-title></titles><pages>1-35</pages><volume>50</volume><number>RR13</number><dates><year>2001</year></dates><publisher>Department of Health and Human Services</publisher><urls></urls></record></Cite></EndNote>(German, Lee et al. 2001). All of these attributes were therefore assessed for the legionellosis surveillance system of Allegheny County. With a focus on the most lacking attribute of data quality, new variables were created and collected for all health-care exposed cases during the 2003-2013 period to indicate the amount of missing data in each case investigation report and risk factor questions section. This analysis was performed to assess completeness of the PA-NEDSS risk factor variables and to elicit recommendations on how data can better be collected for later analyses such as the one intended for this legionellosis project. First, an ‘environmental testing’ variable was created by reviewing the case notes for any comment about environmental testing being done. This variable was an indication of any sort of notes on testing included in the investigation reports, not whether testing was conducted or not, for each case. Next, a ‘remediation recommendations’ variable was created by reviewing the case investigation notes of all cases for any notes about remediation techniques. The following ‘nosocomial risk factor’ variable was created by reviewing the three nosocomial variable questions for an answer (if no answer, missing) and also the correct answer (if wrong answer, misclassification) as determined by the case investigation notes. Finally, an ‘other risk factors’ variable was created for each case by classifying the risk factor answers (outlined more in Appendix) on completeness based on the risk factor questions section and case investigation notes. resultsThe following table outlines the characterization of data quality, specifically completeness of the variables of interest in the investigation reports. The information of interest from the surveillance form included those relating to water testing, remediation recommendations, and nosocomial variables. The variables have both the total number and percent, with the overall total listed at the top of the table.Table SEQ Table \* ARABIC 2. Missing data on variables for health-care exposed cases(Total N=132)Missing (n)Percent Missing (%)Environmental TestingMissing7053.0Listed, but no results (of n=64)2438.7Remediation RecommendationsMissing6247.0Listed, but no details (of n=70)2535.7Nosocomial Risk FactorsMissing1612.1Misclassified6650.0Other Risk Factors (See Appendix)Missing or Incomplete4231.8Misclassified64.5In Table 2, missing is defined as a complete lack of information for a given case. In the ‘environmental testing’ category, listed, but no results is defined as having information about environmental testing, but with no results (i.e. positive or negative cultures) being provided for a given case. In the ‘remediation recommendations’ category, listed, but no details means remediation techniques were discussed with property owner, building manager, etc. but no information regarding what specific remediation technique was recommended was provided for a given case. The ‘nosocomial risk factors’ category represents three variables: not nosocomial, probably nosocomial, and definite nosocomial. In this category, misclassified is defined as an incorrect specification of one or more variables for a given case. Since the three variables are mutually exclusive, misclassified often meant that two or all three variables were selected (as being true); alternatively, it meant that the variable selected was not accurate according to case notes. In the ‘other risk factor’ category, missing or incomplete is defined as either a complete lack of information or partial/incomplete information for all risk factor variables. In this same category, misclassified is defined as one or more variable not being accurate according to case notes. For example, this could mean that the case notes indicated a given case travelled or vacationed during their incubation period, but the overnight travel variable was not selected (as being true).After assessing the CDC attributes of a surveillance system, not all attributes were analyzed to the same extent. There were three found to be most problematic and will be discussed more in Section 4.2. These three attributes are data quality, simplicity (of the form), and acceptability (to the clinicians and public health workers involved). The following attributes were determined either to be irrelevant or to not have an effect on the surveillance system. They are flexibility, sensitivity, predictive value positive, representativeness, timeliness, and stability. Flexibility: Since all forms can be edited at any point in time once they are opened in the PA-NEDSS system, flexibility was determined not to be an issue. Sensitivity: Using the first definition of sensitivity, the proportion of cases is impossible to know without further study of legionellosis, so it becomes irrelevant for this project. Using the second definition, the PA-NEDSS system is definitely able to determine outbreaks or changes in case frequency when health-care facilities and others are actually reporting the cases they diagnose. Predictive Value Positive: For legionellosis, this is only a relevant measure in regards to the accuracy of the diagnostic test, not the accuracy of the surveillance system, so is therefore considered to be outside the scope of this project. Representativeness: For this legionellosis surveillance system, it would be a measure to show how similar in characteristics the reported cases are to all occurring cases. This again cannot be measured by this project since the ability of reported cases to demonstrate the correct proportion and characteristics of all cases would need to be determined by additional epidemiology studies. Timeliness: For the surveillance system, it is theoretically instantaneous if there is staff available to perform any sort of summary or analysis on the surveillance data. Stability: This attribute is not a problem for this surveillance system since the forms in PA-NEDSS are all electronic, and anyone who needs to could access them. Both the system and the definitions (for cases and variables) are unchanging over time.DiscussionAccording to Table 1, 15.8% of reported cases were health-care exposed cases. There is no reference to say if this proportion is to be expected or not, since the CDC did not report a health-care associated or exposed case rate or proportion. Among those cases within the county, however, cases were twice as likely to occur in an acute care facility as a long-term care facility. Health-care exposed cases were also slightly (9.7%) higher than the number of travel-associated cases, and 680-750% higher than the number of health-care facility employee cases, dental work cases, and senior high rise exposed cases. The sex distribution was expected based on US rates, with an approximate 1:2 ratio of female to male cases in both Allegheny County and the US. The age distribution is slightly older than expected based on the national distribution. While more than 60% of cases occur in people ages 60 and up in Allegheny County, only 50% of cases occur in that age group overall in the US. For cases in people 70 and up, the Allegheny County proportion is almost 40%, while the US proportion is almost 30%. This difference continues in the 80 and older group with around 20% in Allegheny County and 13% in the US.Despite the amount of missing data for various risk factors and other information, it is interesting to note how common cases are in long-term care facilities in Allegheny County. It is estimated that 7,000-10,250 elderly persons are living in long-term care facilities each year in Allegheny County ADDIN EN.CITE <EndNote><Cite><Author>Absher</Author><Year>2014</Year><RecNum>24</RecNum><DisplayText>(Absher 2014)</DisplayText><record><rec-number>24</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1398197582">24</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author>Clare Absher</author></authors></contributors><titles><title>Nursing Home Inspector</title><secondary-title>Allegheny County, PA</secondary-title></titles><volume>2014</volume><dates><year>2014</year></dates><publisher>CarePathways</publisher><urls></urls></record></Cite></EndNote>(Absher 2014). Given the number of cases, the estimated rate for legionellosis in long-term care facilities in Allegheny County ranges from 410-600 reported cases per 100,000 people. For comparison, the general elderly rate (65 and older) in the county is 17.3 per 100,000 people, and the elderly rate (60 and older) in the US is 2.3 per 100,000 people. This difference is, at least in part, because of the increased susceptibility of the residents and the increased rate of nosocomial infections occurring, but this only proves the need for determining the most effective remediation and prevention methods and putting them to use in these facilities.Surveillance System ProblemsAs can be seen by Table 2, the various categories in water testing, nosocomial, remediation recommendations, and other risk factors are all higher than 30%, which was determined for this project to be the point when analysis would be ineffective in making any reliable conclusions. Note: while the proportion missing for nosocomial factors is only 12.1%, the proportion misclassified is 50%, which makes the data for that variable very invalid and unreliable. Because of these high proportions of missing, misclassified, or unspecific information among these variables of interest, the ‘data quality’ of the surveillance system comes into questions and this will be discussed more below.This problem points to greater implications for the effectiveness and reliability of the surveillance system overall. While PA-NEDSS succeeded in measuring overall burden of a disease and detecting changes in rates of legionellosis for Allegheny County, the missing data of the surveillance system rendered PA-NEDSS not useful in secondary goals of surveillance systems, such as to monitor disease trends (such as identifying the variables associated with a person getting the disease), to prioritize health resources, to guide in planning and evaluation of programs to prevent and control disease, to evaluate public policy and to provide a basis for epidemiologic research ADDIN EN.CITE <EndNote><Cite><Author>German</Author><Year>2001</Year><RecNum>21</RecNum><DisplayText>(German, Lee et al. 2001)</DisplayText><record><rec-number>21</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1396474810">21</key></foreign-keys><ref-type name="Report">27</ref-type><contributors><authors><author>Robert R. German</author><author>Lisa M. Lee</author><author>John M. Horan</author><author>Robert L. Milstein</author><author>Carol A. Pertowski</author><author>Michael N. Waller</author></authors></contributors><titles><title>Updated Guidelines for Evaluating Public Health Surveillance Systems</title><secondary-title>Morbitity and Mortality Weekly Report</secondary-title></titles><pages>1-35</pages><volume>50</volume><number>RR13</number><dates><year>2001</year></dates><publisher>Department of Health and Human Services</publisher><urls></urls></record></Cite></EndNote>(German, Lee et al. 2001). The implications of the proposed analysis for this project would have been to determine what risk factors are of interest and are most effective for targeted or specific prevention in the county. If for example we had found that performing environmental testing at a facility lead to a 40% reduction of another case occurring at that facility or that of facilities who tested positive for Legionella, those who were told to do a heat and flush (vs. doing chlorination treatment or installing a silver-copper ionization system) were 50% less likely to have another Legionnaire’s disease case at their facility, recommendations would have been made for the county as to whether environmental testing seemed to be effective or what the “best” tested method for remediation was. Since the analysis was not carried out, review of the surveillance system and subsequent recommendations are made instead.RecommendationsThe first intractable attribute was the data quality, specifically completeness, of the surveillance system. Since the data quality was so poor that there was not enough information collected by the Allegheny County Health Department via PA-NEDSS to answer the intended questions, it must be asked if collecting this incomplete data for certain variables is even worth the time and effort. If essentially nothing can be done with the data that was collected (due to the amount of missing data), then the ACHD should strive to either devote more people and time to properly filling out these reports or remove certain risk factor, diagnosis, or other types of questions in the legionellosis form. This second option would leave just the essential questions and sections, which would streamline this form, making it easier and quicker to complete. Obviously, the current questions being asked are useful when their answers are present in large proportions (usually epidemiologists prefer data around 70% complete or not missing). But, if trying to complete non-essential information is draining the health department of its resources and time, then a cost analysis might need to be considered. In the end, the more accurate and complete information the county has on its diseases and the characterization of those with diseases, the better.One recommendation for the Allegheny County Health Department is to become more efficient and accurate when completing each case investigation report for all diseases using their current PA-NEDSS forms. Anyone filling out the reports needs to be properly trained in incubation times, medical terms, and basic surveillance definitions in order to help cut down on misclassifications. One of the biggest problems with the nosocomial variable (and most likely the reason for the large proportion of misclassified data) was that the investigator seemed to use their own discretion (e.g., when deciding between the definite, probably, or not nosocomial options) rather than strictly using the definitions. It should be stressed upon these investigators that the definitions exist to not only classify the cases individually, but to keep the cases within each category comparable. And lastly, investigators should understand that negative environmental testing results does not mean a facility is not colonized with Legionella. Negative results could also mean that the lab or testing specialist did not sample the right water sources or surfaces, did not properly collect or collect enough water, or got a false negative lab culture result.As part of potential PA-NEDSS form improvements, the attribute of simplicity of the form should be address and improved upon as well. To make things easier for investigators to complete certain questions, the onset dates or at least an estimated window of dates should be required for all forms as to allow for better indication of exposures during incubation times. Furthermore, all of the risk factor variables should be classified as occurring during the known or estimated incubation period. Currently, the use of an incubation period seems to be used sporadically by investigators filling out the risk factor questions and this makes case reports inconsistent. Additionally, those in charge of surveillance for the county need to prioritize the variables that need answered. If there does not seem to be enough time to answer all questions in most reports for a certain disease, then the questions of most importance should be determined and targeted so that at least some variable questions are completed enough as to allow for future analysis and research for the county. This could be done by integrating a skip pattern or required fields into the electronic form so that these essential questions will have answers for each case. Another recommendation for more accurate and characterized understanding of legionellosis and its risk factors are research studies and standardized policies. First, overall studies of legionellosis (such as case control or cohort studies) could help the community better understand what risk factors are the best predictors of legionellosis occurring. Second, similar studies could be very useful for determining usefulness and effectiveness of remediation techniques. At the very least these various techniques need to be better analyzed so that a (national or worldwide) standard policy could be created as to which remediation techniques are actually effective and in what situations. Third, it would behoove the scientific community to appeal for better diagnostic practices on the part of clinicians. Currently, most clinicians treat pneumonia empirically. In other words, doctors do not routinely test for the type of pneumonia a patient may have, but rather they will start them on a course of antibiotics that would clear most types of pneumonia. However, if this continues as the standard practice, the larger community may never know the true proportion of pneumonia caused by Legionella (and other bacteria for that matter). It become much more difficult, if not impossible, to create effective prevention programming and policy when the disease characteristics and its frequency are not understood.A broader recommendation is better cooperation between public health and health-care workers. This would increase the acceptability of any changes to the overall surveillance system for all involved. This third attribute, acceptability, is defined as the willingness of persons involved to participate in the surveillance system. Since legionellosis is a reportable disease, it makes the most sense to discover and report as many cases as possible across the US. In other words, there is a definite use and necessity for the current electronic surveillance system. What should be considered is both acceptability by clinicians and public health workers. The case reporting process usually starts with the clinicians who are diagnosing and reporting the disease first hand. When cases go undiagnosed (and therefore unreported) it not only makes the US or local rates deceivingly low, but there could be some bias occurring with those that are diagnosed, which makes the characterization of cases less reliable and accurate. In other words, this gap in all cases and reported cases makes our knowledge incomplete as well. For instance, we don’t know with certainty if the age, gender, and risk factor proportions reported in say Table 1 are accurate descriptions for all cases or just the reported cases for Allegheny County. This gap between all cases and reported cases is not acceptable to the public health investigators and makes it hard for public health workers to create policy around Legionella prevention and/or remediation and also to prevent future cases from occurring. This is largely important because, unlike other forms of pneumonia, Legionella is not communicable. Since legionellosis is not spread from person to person, cases can only be prevented at their original sources ADDIN EN.CITE <EndNote><Cite><Author>CDC</Author><Year>2013</Year><RecNum>1</RecNum><DisplayText>(CDC 2013)</DisplayText><record><rec-number>1</rec-number><foreign-keys><key app="EN" db-id="2rx20rtxh0wer9eztxhp2epg9sfwr5ttsefp" timestamp="1395349901">1</key></foreign-keys><ref-type name="Web Page">12</ref-type><contributors><authors><author>CDC</author></authors><secondary-authors><author>Centers for Disease Control and Prevention</author></secondary-authors></contributors><titles><title><style face="italic" font="default" size="100%">Legionella</style><style face="normal" font="default" size="100%"> Home Page</style></title></titles><volume>2014</volume><number>January</number><dates><year>2013</year><pub-dates><date>February 5</date></pub-dates></dates><pub-location>Centers for Disease Control and Prevention</pub-location><publisher></publisher><urls><related-urls><url>;(CDC 2013). Therefore, if only for a temporary period, clinicians should start testing and diagnosing the specific causal agent in all cases of pneumonia until a better understanding in etiology, prevalence, and control of legionellosis infections can be achieved.APPENDIXTimeline of Risk Factor VariablesStarting in 2003, the following risk factor variables were added to the PA-NEDSS legionellosis form: ‘travel overnight’, ‘dental/oral surgery’, ‘visited or stayed in a health-care facility’. The nosocomial variables (definite, probably, and not nosocomial) were also added at this time.Starting in June of 2004, the variable ‘stay or work in a hospital or health-care facility’ was added.Starting in 2012, a variable called ‘day trips’ was added. This was defined as a case travelling, but without overnight stay.Starting in February of 2013, the following variables were added to the risk factor portion of the legionellosis form: ‘used whirlpool’, ‘used respiratory equipment’, ‘previous or current smoker’, ‘was exposed to water mist’, and ‘lived in long term care or assisted living facility’.bibliography ADDIN EN.REFLIST Absher, C. (2014). "Nursing Home Inspector." Allegheny County, PA. 2014.ACHD (1997). Approaches to Prevention and Control of Legionella Infection in Allegheny County Health Care Facilities. Pittsburgh, Allegheny County Health Department.ASHRAE (2013). Prevention of Legionellosis Assiciated with Building Water Systems. Atlanta, GA, ASHRAE. BSR/ASHRAE Standard 188P.CDC (2005). Procedures for the Recovery of Legionella from the Environment. U. S. D. o. H. a. H. Services. Atlanta, Georgia, Centers for Disease Control and Prevention.CDC (2011). Legionellosis-United States 2000-2009. Morbidity and Mortality Weekly Report, U.S. Department of Health and Human Services. 32: 1083-1086.CDC (2013, February 5). "Legionella Home Page." Retrieved January, 2014, from , A. A., et al. (2001). "Etiology of Severe Pneumonia in the Very Elderly." American Journal of Respiratory and Critical Care Medicine 163(3): 645-651.Fields, B. S., et al. (2002). "Legionella and Legionnaires' disease: 25 years of investigation." Clin Microbiol Rev 15(3): 506-526.German, R. R., et al. (2001). Updated Guidelines for Evaluating Public Health Surveillance Systems. Morbitity and Mortality Weekly Report, Department of Health and Human Services. 50: 1-35.Harris-Kojetin, L., et al. (2013). Long-term care services in the United State: 2013 overview. Hyattsville, MD, Center for Health Statistics.Howden, L. and J. Meyer (2011). Age and Sex Composition: 2010. 2010 Census Briefs, US Department of Commerce-US Census Bureau.Kokenda, J. (2013). Allegheny County Census Data. Allegheny County Health Department. (2012). "Guidelines." 2014, from (2000). Report of the Maryland Scientific Working Group toStudy Legionella in Water Systems in Healthcare Institutions. Baltimore, Maryland, Maryland Department of Health and Mental Hygiene.Musa, D., et al. (2003). The State of Aging and Health in Pittsburgh and Allegheny County, University of Pittsburgh Graduate School of Public Health.Neil, K. and R. Berkelman (2008). "Increasing Incidence of Legionellosis in the United States, 1990–2005: Changing Epidemiologic Trends." Clinical Infectious Diseases 47(5): 591-599.NIH (2014, 7 March 2014). "Legionnaires' Disease." 2014, from , A. (2005). Hospital-associated Legionellosis. Albany, NY, State of New York Department of Health.SAHealth (2008). Guidelines for the control of legionella in manufactured water systems in South Australia, Government of South Australia, Health Protection Programs, Public Health Services, Public Health and Clinical Systems, and Department for Health and Ageing.Seenivasan, M. H., et al. (2005). "Legionnaires' Disease in Long-Term Care Facilities: Overview and Proposed Solutions." Journal of the American Geriatrics Society 53(5): 875-880.Silk, B. J., et al. (2013). "Legionnaires' disease case-finding algorithm, attack rates, and risk factors during a residential outbreak among older adults: an environmental and cohort study." BMC Infect Dis 13(1): 291.Stout, J. (2013). Planning Meeting for Legionella Paper. Allegheny County Health Department. L. Hamitlon.Torso, L. (2014). Problems with PA-NEDSS. University of Pittsburgh Graduate School of Public Health. L. Hamilton.Trop Skaza, A., et al. (2012). "Epidemiological investigation of a legionellosis outbreak in a Slovenian nursing home, August 2010." Scand J Infect Dis 44(4): 263-269.WHO (2007). Legionella and the Prevention of Legionellosis. J. Bartram, Y. Chartier, J. V. Lee, K. Pond and S. Surman-Lee. India, World Health Organization. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download