American Colon Cancer Society

[silence] welcome to part 2 of screeningfor colorectal cancer: optimizing quality, the cdc-sponsored programemphasizing the critical importance of improving qualityin order to more effectively reduce colorectal cancer incidence and mortality. in part 2, we're going to focus on delivering high-qualitystool blood testing in primary care. i'm dr. richard wender,the chief cancer control officer at the american cancer society

and a member of the facultythat had the opportunity to put this program together. so here's what we're going totalk about in part 2. we're really going to focuson stool blood testing on the importanceof offering a stool blood test, and i think the data are compellingof how important it is if we're really going to increasescreening rates across the nation. we're going to talk about how to developa high-quality stool testing program, and any of you who've done thisknow it is not as simple

as just handing a patient a kitand saying, going home and do this. you know that if that'sthe approach that's taken, you have at best a 50% chanceof ever seeing that kit again. and we also haveso many instances across the nation where the testingis not collected correctly. it can't be developedor patients who have a positive test who then do not get the very importantnext step of colonoscopy, which is really the step that makesa difference finding pathology. and we're going to go throughwhat are the components

of a high-quality stool testing program,including identifying the right patients, making sure you've accountedfor the risk of the patient, and then workingwith your team in your office around communication,around instructions for patients, around navigation,around test completion, around processing the testand making sure that we follow upon abnormal results. so let's start with a pop quiz to get your brainflowing here and working.

why is it importantto offer stool blood testing as an option for screening? so give that some thought,and let's talk about it. as all of us know, and the dataare clear in the united states, colonoscopy is now the most frequentlyused screening test for colorectal cancer. however, when it'sprovided every year, annually, to average-risk patientswith the appropriate follow-up, stool occult blood testingwith high-sensitivity tests can provide similar reductions in mortalitycompared to every-ten-year colonoscopy,

and you also seesome reduction in incidence because patientswith advanced polyps are more likely to have microscopicbleeding and have a positive test. there are some real notedadvantages of stool blood testing, and these are advantagesthat matter to a lot of patients. well, first, it's less expensive. it can be offeredby any member of the health team. it requires no bowel preparation. it can be done totally in privacyat home no time off from work.

you don't need to haveanother person available to drive you home after a procedure,and, of course, it's noninvasive. it has no riskof causing pain, bleeding, bowel perforation,or any other adverse outcome, and, of course,colonoscopy is required, but only if the stool bloodtesting is abnormal, so if a patient has ten annualnegative tests for blood in their stool, they will not needa colonoscopy at all. i think even more criticallyare the data we have

about patient preferenceand patient action. here's a study of patient preference--in fact, two studies-- one, a diverse sample of 323 adults who were givenside-by-side descriptions of fobt versus colonoscopy, and half of the patients preferred fobt, and half of those felt very stronglyabout that preference. two hundred and twelve patientsat four health centers in texas rated different screening options

based on the attributesthat were described. here, about a thirdpreferred colonoscopy, and about a thirdpreferred stool blood testing. a nationally representative sample of over 2,000 veteransadministration patients given brief descriptionsof each modality, and we saw similar results. a little over a thirdpreferred colonoscopy, and a little less than a thirdpreferred fobt.

five hundred and twenty-fiveracially diverse average-risk patients at two safety-net ambulatorycare sites in boston were given actually an interactivedecision aid for crc screening that was carefully developedby a research team, and here, the preferencescame out somewhat different. actually, 60% preferred colonoscopy,but still, just under 30% preferred fobt. interestingly, in this study in boston,although patients preferred fobt, they were far less likely to havethat preferred test ordered than those who preferred colonoscopy,

and they were more likelyto have no test ordered at all, 31% of that group versus 15% for the groupwho preferred colonoscopy. so we have some work to do, but i think this studymay be the most important of all. this was a randomized clinical trial in which about a thousandethnically diverse patients in san franciscocommunity health centers received differentrecommendations for screening.

colonoscopy was recommendedin about half, stool testing in about half. well, interestingly,only 38% of people who got a recommendationfor colonoscopy actually completed the test,whereas for that group who had stool blood testingrecommended, 67% completed. i think it justbrings home the point. we are not going to getto an "80% by 2018" goal. we're not going to get the enormouspotential of stool blood--

of colon cancer screening,unless we include both options. the reality is,some patients simply forgo any colorectal cancer screening if they're not offereda home stool blood test as an alternative to colonoscopy, and in fact, as we've saida number of times now, the choice of an annual stool blood test is quite a reasonable optionfor average-risk patients, and if successfully implementedover a ten-year period,

we will avert about the same numberof colon cancer deaths. the one thingi want to drive home, though, and i really put into placein my own practice, and i've been using fit,fecal immunochemical testing, as an important alternativefor years now-- patients need to understand,when they select a stool blood test option, they're making a commitmentto two things: one, they're making a commitmentto an annual test as opposed to a once-every-five- or ten-year test,

and number two, most importantly,they have to understand that if the stool blood test is positive,there's really no longer any choice. you must have a colonoscopy, and i've heard storiesin primary care which worry me, where patients hada positive stool blood test but were still essentiallythen offered colonoscopy as an option as opposedto a very clear and firm and persistent recommendation. so when you offer stool blood testing,make sure patients understand it's annual,

and if it's positive,they need a colonoscopy. so i think, to maximize the numberof patients who get screened, we must include stool blood testas a screening option. emphasizing that thisis for average risk is critically important, and we're going to revisit thata number of times. in part 1, we went over the risk criteria,and you'll see that for people at higher risk, stool blood testingis generally not recommended, with one exception, for the"very slightly increased risk" group. this means that all practiceneeds to develop systems of care

to make sure that we identify the patientswho are eligible for stool blood testing and that if we choose that option,we make sure we have a system in place to implement annual screeningat appropriate time intervals with high-quality instructionsand attention to the follow-up. i think one of the mostheartbreaking instances of all is the patient who hasa positive stool blood test but somehow it got lostbecause the system wasn't in placeto ensure follow-up. so there's a lot moreto stool blood testing

than just handing out a kit, as i said in the introduction,and the next set of slides, we're going to go over these steps, and there's a wonderful resourcethat was produced by the nationalcolorectal cancer roundtable with support from acs and cdc. it's on the link on your screen, and please find some timeto go through that resource. it's very comprehensive

and provides far more detailabout how to implement the steps that we're about to review,but here are those steps that it takes to implementa high-quality program. one, you need an effective test. two, you have to identifythe eligible patients. three, you need a trained staff who knows how to communicatewith patients effectively. four, you need appropriatetest instructions. five, you need steps to ensurehigh test-completion rates,

and you need to measurecompletion rates. number six, you needa high-quality test processing either in the office or through a lab. seven, you need follow-up of abnormaltest results with colonoscopy. and eight, you need to make sure the steps are in placeto get this done every year. here's our second pop quiz. is screening with a standardguaiac-based test like hemoccult ii a good way to screenfor colorectal cancer?

well, step one is to selectan appropriate test, and obviously,it's a very important consideration when putting into this kind of program, and the thing we paythe most attention to is the sensitivity of the test. in low-resource nations,we choose a different sensitivity, but in the united states, we believe that choosinga highly sensitive test is the best fit for the resources

and what we're trying to accomplishin the united states. so the guideline that was producedby the american cancer society and by the gi groupsand the radiologists made a decisionto recommend only tests that had a sensitivity for cancerwith one-time use of at least 50%. that means that we would includehigh-sensitivity guaiac-based tests and fecal immunochemical tests that have been testedwhere we know sensitivity. but older tests,standard tests like hemoccult ii

actually would not meetand do not meet that 50% threshold. i do want to emphasize that this 50% thresholdis somewhat arbitrary and was recommendedby this guideline team, again, placing high valueon having sensitivity be high in order to havethe biggest impact on mortality. so as i said, standard guaiac testsdo not meet that 50% threshold, and the united statespreventive service task force made a very similar statement,

although they didn't specifically usethe 50% threshold. they specifically saidthat lower-sensitivity tests should not be used. so bottom lineis that neither the task force nor the multisociety/acs guideline recommend using the older testslike hemoccult ii or tests with a sensitivitythat is particularly low. so what are the high-sensitivity tests? there are high-sensitivityguaiac-based tests

that exceed the sensitivity threshold. hemoccult sensa is a great exampleof such a high-sensitivity test with a sensitivity for cancerof 64% to 80% with a specificityof about 87% to 90%. so that absolutely does meetthat 50% threshold. so let's talk a little bitabout high-sensitivity guaiac tests. they require stool samplescollected on three different days after up to seven daysof dietary and medication restrictions, which i knoware familiar to most of you.

each stool specimen is collected by usinga collection stick to take samples from two different areas of stoolfrom each bowel movement. the stool should be collectedon the paper provided floated on top of toilet water before it comes into contactwith the water itself. the kit is generally manually processedin the clinic or in a lab. it's a very affordable test-- average medicare reimbursementof $4.48 per test. the other widely availableappropriate option

are fecal immunochemical tests, but it's important to realizethat if you've seen one fit, you've seen one fit. they're all collecteda little bit differently. they have a little bit differentperformance characteristics, and in choosing a test,it's your responsibility to actually know what the performancecharacteristics of these tests are. but here's the good news. fits in general are easier to use

and more likelyto be completed by patients than the guaiac tests,and why is that? first, there's no needfor dietary or medication restriction, and second,most fit tests require samples from only one or twoconsecutive bowel movements. so fewer bowel movements,no dietary or medication restrictions, they're easier to use,and in fact, there are some data from a number of the testsshowing better return rates. one drawback to the fitis that they are slightly more expensive

than the guaiac tests, but generally,it is possible to make arrangements to reduce or eliminate up-front costsfor clinical practices where the lab actually does the chargingonce the test is returned. fits are usually reimbursedat a higher rate than guaiac. medicare reimbursement rateis $21.86 for each completed test, so still very affordablecompared to a test like colonoscopy but a bit more expensivethan the guaiac tests. as i mentioned earlier,not all fits are the same. different fits use differentcollection methods

in requiringvarying numbers of samples, so you need to understandthe test you're using. some can be processed in the office--point-of-care tests-- but others cannot, and some fitsuse a machine or analyzer to measure hemoglobineither in the office or sent to a lab. also important to understandthat different tests have different hemoglobincutoff values to define an abnormal test level. lower cutoff values, of course,increase sensitivity but lower specificity,

and if you use these lowerhemoglobin cutoff values, you'll get a more sensitive test and a higher numberof people who test positive and will need a colonoscopy. but in general, i think,in the united states, the more sensitive stool testsmake more sense for our resources. in making these selections, one of the tough dilemmasthat you do face is that not all these fitshave been tested rigorously,

so we recommend using a fit that has been evaluatedin clinical practice and for which data on performancein peer-reviewed literature show at least a 50%sensitivity for cancer. here's some data to help youmake that decision. a recent systematic review comparedresults of studies on different fits, and they looked at studies donein average-risk, asymptomatic patients with an appropriate reference standard, which means colonoscopyor greater than two years of follow-up

to see what happenedto these patients. those were the standardsthat were used to see what the patientsactually ended up having. this included several studiesof polymedco fits, one study of hemoccult ict, and studies of fits that are notavailable in the united states. the range of sensitivityfor cancers was 56% to 100%, and range of specificityfor cancer of 83% to 97%. it excludes studies of fitsthat have been discontinued.

obviously, that wouldn'thelp us very much. so at this time, the brand of fitthat has been most extensively tested and is available in the united statesis called oc fit-chek, and it's made by polymedco. let me tell youa little about this test. it's provided as a one-sample kitin most cases. that's good news for patients,just one bowel movement. the collection method involves inserting the probeseveral times into the stool

to a point on the probejust above the ridges. and i recommend,if you've never used the test, just take out the probeand look at it. it's very easy to understandwhat we mean by the ridges and then placing the collection probeinto a small tube. just like the guaiac test,it's better to do this test on a stool that's floating on paperthat sits on top of the water, rather than doing the test on a stoolthat's sitting in the toilet water itself. two ways to process this test,

and there are tools availablefrom the company to do it. there's manual oc-light. this is a point-of-care assay, and it has an estimated sensitivityfor cancer reported at around 93%. the other approach is automated,oc-auto test, which uses an automated analyzer,and it has a very similar sensitivity, reported around 86% or 87%. very similar to the manual test. so here's the next pop quiz.

is performing a stool blood testusing a stool sample collected during a digital rectal exama good way to screen your patients? well, although this practiceremains very common, it is ineffective and not recommended for screening. all stool collectionfor colorectal cancer screening should be done at home. the stool collectedon a digital rectal exam may not be sufficientor sufficiently representative of stool collectedfrom a complete bowel movement,

and there's actually no evidencethat any type of stool blood testing is sufficiently sensitivewhen collected from a stool sample during a digital rectal exam. therefore, high-sensitivityguaiac fobt and fit should be completedby the patient at home and not an in-office test. there are some physicians,and i understand this, who still do a rectal examand believe in it, and that's perfectly fine.

we're not telling youto stop or to start doing it. we are saying that--understand that that patient has not been screenedfor colon cancer. to be screened for colon cancer, they need to have oneof the recommended tests, which includes the optionof doing this fit or high-sensitivity guaiacfobt collected at home. step two, identifyingeligible patients. so, obviously, if you can useelectronic patient data, do it.

that obviously will help youidentify the right people who need to be screened. the primary targetare average-risk adults, and if you need some refresheron what defines average risk, we went over that in detail in part 1. but average-risk adults age 76 to 85 can be considered as wellon a case-by-case basis, which we also talked about. if you're going to screen the elderly,

it's much more helpfulto screen people who are not up-to-date,who've not had a recent test. so average risk, 50 to 75,and then on case-by-case, 76 to 85. many patients at higher-than-averagerisk should be screened with colonoscopy rather than any formof stool-based testing. next pop quiz: should you recommendan interim stool blood test to an average-risk patient who had a normal colonoscopyseveral years ago?

well, in fact, there's no evidence for performing an interimstool blood test for average-risk patients who've had a negativehigh-quality colonoscopy within the preceding ten years, and i still see this being donequite frequently. either the patient doesn't understandor the clinician doesn't understand that if you've hadthat normal colonoscopy, that patient is considered up-to-date,fully screened, for the next decade. risk assessment is always critical.

we should not be screeninghigher-risk patients with a stool blood test. so outreach programsthat use stool blood testing need some system to identify"higher than average risk" patients to make surethey get colonoscopy when it's indicatedrather than stool blood test, and that takes some work.it's really the responsibility of the screening centerto have such a system. step three, training staffto communicate with patients.

you know, if you givea stool blood test to somebody and you do not explain explicitlyand clearly what it's for, how often it needs to be done,and how it's done, you shouldn't be surprisedthat patients don't return it. they're going to open it up,and they're not going to believe that that's really whatyou intended for them to do, so you have to bevery explicit with patients: why is this screening important, why high-sensitivityguaiac testing or fit

are useful optionsfor colorectal cancer screening, and that they're great tests. and patients need to understandthe testing interval and the importance of following upfor an abnormal test result. you need to develop, tailor,and practice delivering scripts to generate enthusiasm for screening among staff and patients,so here are some examples. "this easy test can save your life." "this test is not painful.

it can be done in the privacyof your own home." "we want to be surethat everyone in our community has a chance to take advantageof colorectal cancer screening." "all our doctors recommend this test." i like that one. i think it gives a really good message, and all of those are usefulas a way to help patients understand they're not gettingan inferior test. they're getting a great way

to reduce their riskof dying of colon cancer. the whole fieldof screening navigation using patient navigatorsis rapidly changing, and we're learning so muchabout the extraordinary value that patient navigators can bringto ensuring adherence and quality. there are recent studiesthat have demonstrated the effectiveness of patient navigatorsin improving adherence. they can provideculturally sensitive material through people who havethe right kind of training

and know how to do it. individualized assistance--and patientshave an amazing array of questions when it comes to screening, and they can help patientsovercome system barriers such as follow-upto an abnormal result. there are many different thingsthat navigators can do. they can get patientsscheduled for screening. they can do outreach. they can explain the technique

and make sure that patientsget their questions answered. they can assurepatients really do understand. they can ask them to repeat back so that they have confidencethat patients know what's being asked, and they can address patient barriersthat patients confront all the time, such as language issues,transportation, et cetera. one of the other things that we can involveour whole system in is making surethat test results are received

and, even more importantly,that abnormal results get the follow-up that they absolutely need. navigators can help identifythe treatment resources and support networks,particularly for people from low resource. number four,provide appropriate instructions. here your staff is essential. the primary clinician themselves, number one,doesn't need to be doing this, number two, doesn't havethe time to be doing this.

clinicians and staff should reviewmanufacturer instructions carefully. make sure you know how the testthat you're using should be collected, and make sure that that'sconsistently communicated. you may need to adjustthe written instructions that come with the packaging to take into account languageissues, literacy issues. make sure the low-literacyversions are available. simplified translations and visual aidscan be very, very helpful. demonstrating the techniquethrough an illustration

can be hugely helpful. make sure that patientsunderstand the importance of dating the specimen and how to return the specimenpromptly to the laboratory. several practices havedeveloped video instructions, which is a great wayof demonstrating this entire process. it won't surprise you to knowthat language makes a difference, so brief, low-literacy videosin multiple languages explaining the importance of screeningand the general principles

of both guaiac and immunochemicaltest completion have been developed, and they're availableon the internet, so please take a momentto click on this link, and you'll be really pleasedat how many languages have resourcesthat will help your patients. there are other low-literacyand multilingual print materials that are availablefrom kit manufacturers, crc screening advocacy groups,or on the internet. so more help is out therefor you to help your patients

than you might realize. here are some examplesof written instructions for the polymedco fit for english, spanish, chinese,vietnamese, and russian. so make sure thatthey've field-tested this a little bit with their patients they're caring forand get the feedback that these translations are workingand are understandable. number five, ensuring hightest completion rates. there's really a numberof approaches to this,

including both in-reach methods, making sure you catch patientsat the point of care, and outreach methods, making sure you're managingan entire population. i think the most valuable toolfor in-reach today are the electronicpoint-of-care reminders. there are still practices,by the way, who are using manual point-of-carereminders, and they work, not dissuading from that,

but to the extent that we can learnto harness our emrs, the better. involve staff, in additionto the primary care clinician, in the process of identifyingeligible patients and offering the testing. there's some amazing exampleswhere patient-- where staff, even checking in the patient,see a screening reminder that the patient is due for screening, and integrated health delivery systems where that reminder pops up everywherethroughout the system. waiting room videos, posters,brochures could all help,

and the cdc has terrific materialin their screen for life campaign, which offers a varietyof these kinds of materials that you could use in your office. when you provide kits, make sure you do itat every opportunity. patients come infor all sorts of reasons, so you can have a standing order that allows nursesto provide the kits during nurse visits. one of the most fully developed, described,and evaluated programs is flu-fit,

and a principal investigatoris one of our faculty for the flu-fit program-- patients coming in for flu shotswhere you put in the system to make sure they also getan annual reminder for their fit test. any sort of visit offers an opportunityto improve screening rates. the nci has compiled a whole set of research-tested interventionprograms for screening, so take a look at this link,and find a approach that fits your office practice.

so we talkedabout in-reach methods. how about outreach methods? again, electronic health registriesthat are built from your emr can be an extraordinarily valuable tool. mailed invitations,where you either include the test kit or you don't include the test kit, have been shownto increase screening rates, and some practices have personal or automated phone callinvitations to participate.

the more personaland the more frequent your reminder, the more likely you areto get a test done. if you've recommended a testbut it has not been returned, you need a system to implementtelephone or mailed reminders within two to three weeksof receipt of the test kit. there are quite a bit of datashowing that many patients keep that kit right within reachbut do not do it right away, and that little extra pusha couple weeks later can absolutely make a difference.

step six, ensuring high-qualitytest handling and processing. so, if at all feasible,if the patient did not write the date of collection on the sample, find out when that date was--when the sample was actually collectedbecause occasionally-- fortunately, not frequently--patients collect the specimen but don't return it right away. using trained, experienced personnel to develop and report the test resultsis obviously critical.

people have to knowhow to use the equipment to get the right answer. if you send this to a centrallaboratory for processing, i think that's a great strategy because you get back a piece of paperthat says positive or negative. that's what we do in our practice, and i find thata tremendous prompt to action when i get that test result backthat says positive. monitoring your test positivity rates

and investigate if thingsare not happening correctly, particularly if you'redeveloping it in your office. there are ways to reduce the riskof false negative results, and that's by making sure patientsunderstand how the test is collected. make sure you're carefully followingthe manufacturer's instructions. returning the kits promptlyimproves their performance, and delay in return will reducethe sensitivity of the test. make sure the testis stored appropriately. high temperaturesbefore or after completion

contribute to degradationof hemoglobin. interestingly, there aresome seasonal variations in fit positivity rates,probably due to these temperatures, and the good news is, companies are working on waysto improve the buffers to reduce deteriorationat high temperature. well, you know, we keep repeatingthe most important points, and this time, not only did we repeat it,we put it in red because it is so critical. all patients with an abnormalstool blood test

must receive a colonoscopy in order for this testto have any benefit. a positive stool blood testwithout a colonoscopy should not be consideredto be a completed screen. so have a system in place that if a patient tests positive,they get that colonoscopy. one thing i occasionally will tell patientsand staff and other clinicians, the patient with a positive stool blood test is eight times more likelyto have a polyp or a cancer

than someone who does not havea stool blood test. it's a far more stronglycorrelated prediction of an abnormal resultthan even a positive family history. so these patients are at high risk,and they need to be referred. navigation makesa huge difference here. make sure that patients-- you have a way to trackthose abnormals, and be very persistentin tracking them down and making sure they getthat colonoscopy referral made.

collaboration with the endoscopist to ensure prompt and proper follow-upobviously can be a huge help. step eight, ensure annual testcompletion: so, not so easy. you really needsome electronic tools here. it's really hard to do this manually. so registries, outreach systems, mail and telephone,to remind and provide testing to patients who don't come infor a visit every year can really help youget that annual testing done.

as i said at the start,and i'll say it again now, provide consistent messaging about the importanceof annual test completion. it's a point you should be makingwhen the patient first makes the choice to be screenedwith a stool blood test that they've made an agreementwith you to be tested every year. quality screeningis critically important, and as you know so well,we just don't get better at things we're not measuring.

so assess your numbersand rates of eligible patients, the test kits provided,the return rate, the processing, the tests kits that were rejectedby the laboratory because they werereturned incorrectly, your rate of abnormaltest results, and, of course, your rate of completing colonoscopyfor those who test positive. and finally, monitor thatsustained annual participation. that's whatthe clinical trials show. that's what the models show.

we can avert the same numberof colon cancer deaths, assuming one hundred percentcompliance with annual screening. let me put a little bit of extra detailabout this issue of referring for colonoscopyif the test is abnormal. one mistake we occasionally seeis that people don't trust the result, particularly if it wascollected in the office or they think the patientdidn't follow the dietary advice. we want to be really clearin this quality tool that that is not an appropriate step.

whether or not the patientfollowed the dietary advice or changed their medicine,if the test is positive, it should be considered positive, and the patientshould go for colonoscopy. there is no indication to repeateither the guaiac or the fit. the reality is, of course,that a normal result on a repeat would not rule out polyps or cancer. which test would you pay attention to? so do not repeat.

everybody with an abnormalshould have a visual colonoscopy. it's not appropriateto do a ct as follow-up or a flex sig barium enema. these patients who test positiveare at higher risk and should have a full colonoscopy. so let me try to summarizeall of these steps to develop a high-qualitystool blood testing system. one, select an effective testand have patients complete it at home. no testing in the office.

number two, make sureyou have the right patients and that high-risk patientshave been identified and referred for a colonoscopy. number three, work with your staff to make sure that they shareyour enthusiasm about screening and are really goodat communicating with patients. number four,work on the test instructions. fit the language and the culture and get some feedbackfrom your patients

to make sure they really work. number five, do everything you canto ensure high test completion rates and monitor how you're doing. number six,process the tests correctly, either in your office or in a laboratory. number seven,make sure you have a process in place where every abnormal test is detected,found, and steps are in place to get that patient in for colonoscopy. finally, number eight,emphasize and have systems in place

to ensure that the test completionis repeated every year. thank you for viewing part 2,focusing on stool blood testing in our quality colon cancerscreening program. part 3 will focuson your role in improving quality for a colonoscopy-basedscreening program for all of your patientswho undergo colonoscopy, and it'll be delivered by my colleague,dr. david lieberman.