統(tǒng)計質(zhì)控與分析目標(biāo)質(zhì)控方法在臨床化學(xué)檢驗中的應(yīng)用--ppt課件

1、統(tǒng)計質(zhì)量控制和分析目的質(zhì)量控制方法在臨床化學(xué)檢驗中的運用廣東省中醫(yī)院檢驗科黃憲章室內(nèi)質(zhì)量控制方法定性測定工程的室內(nèi)質(zhì)控利用患者數(shù)據(jù)質(zhì)控方法利用患者標(biāo)本質(zhì)控方法控制質(zhì)量控半定量測定工程的室內(nèi)質(zhì)控利用患者數(shù)據(jù)質(zhì)控方法利用患者標(biāo)本質(zhì)控方法控制質(zhì)量控定量測定工程的室內(nèi)質(zhì)控利用患者數(shù)據(jù)質(zhì)控方法患者結(jié)果均值法正態(tài)均值法，挪動均值法差值檢查法Delta-核對法患者結(jié)果多參數(shù)核對法血型陰離子隙(Anion Gap)法酸堿平衡法臨床相關(guān)性研討利用患者標(biāo)本質(zhì)控方法患者樣本雙份分析保管患者樣本的檢測留樣再測患者樣本測定的方法學(xué)比較同一工程不同方法一樣方法不同檢測系統(tǒng)控制質(zhì)量控提 綱CNAS運用闡明草案對室內(nèi)質(zhì)控的要

2、求統(tǒng)計質(zhì)量控制分析目的質(zhì)量控制室內(nèi)質(zhì)控數(shù)據(jù)的管理CNAS運用闡明草案對室內(nèi)質(zhì)控的要求臨床化學(xué)臨床免疫學(xué)臨床血液學(xué)臨床體液學(xué)臨床微生物學(xué)基因擴增檢驗臨床化學(xué)5.6.1實驗室應(yīng)制定室內(nèi)質(zhì)量控制程序，可參照GB/T20468 -2006，內(nèi)容包括：運用恰當(dāng)?shù)馁|(zhì)控規(guī)那么，檢查隨機誤差和系統(tǒng)誤差；質(zhì)控標(biāo)本的類型、濃度和檢測頻度；應(yīng)經(jīng)過實驗室實踐檢測，確定精細(xì)度質(zhì)控品的均值和規(guī)范差；改換質(zhì)控品批號時，宜新舊批號平行測定，獲得20個以上數(shù)據(jù)后，重新確定新批號質(zhì)控品的均值；繪制室內(nèi)質(zhì)控圖，可運用Levey-Jennings質(zhì)控圖和/或Z分?jǐn)?shù)圖。質(zhì)控圖應(yīng)包括質(zhì)控結(jié)果、質(zhì)控品稱號、濃度、批號和有效期、質(zhì)控圖的中心

3、線和控制界限、分析儀器稱號和獨一標(biāo)識、方法學(xué)稱號、檢驗工程稱號、試劑和校準(zhǔn)品批號、每個數(shù)據(jù)點的日期和時間、干擾行為的記錄、質(zhì)控人員及審核人員的簽字、失控時的分析處置程序和糾正措施等。實驗室應(yīng)檢查失控對之前檢驗的影響。臨床免疫學(xué)5.6.1 實驗室質(zhì)量控制程序應(yīng)符合如下要求：質(zhì)控標(biāo)本的類型宜選擇人血清基質(zhì)，防止工程菌或動物源性等的基質(zhì)、濃度陽性質(zhì)控物濃度宜2CUT OFF 值左右，陰性質(zhì)控物濃度宜0.5CUT OFF 值左右和位置不能固定而應(yīng)隨機放置、穩(wěn)定性宜選擇消費者聲明在一定保管條件下如2-8或-20以下有效期為6個月以上、分析頻率的選擇應(yīng)滿足臨床要求的分析范圍的測定；用數(shù)值斷定結(jié)果的工程如E

4、LISA、發(fā)光技術(shù)等可運用Levey-Jennings 質(zhì)控圖。質(zhì)控圖宜包括以下信息：分析儀器稱號和獨一標(biāo)識、方法學(xué)稱號、檢驗工程稱號、試劑和校準(zhǔn)液批號、質(zhì)控品稱號、批號和有效期；橫坐標(biāo)X軸每個點闡明的是分析批或檢測日期，當(dāng)檢測日有多個批次時都應(yīng)標(biāo)出；縱坐標(biāo)Y軸用吸光度值或含量點圖，至少要引出中心線質(zhì)控物均值和上下失控限三條線；中心線為質(zhì)控物均值；利用cut off驗證值可以直接確定上下失控限；陰、陽性質(zhì)控物的測定值均應(yīng)在質(zhì)控圖中標(biāo)出。只需當(dāng)質(zhì)控物批號改動時才重新繪制新的質(zhì)控圖，不能隨試劑批號的改動而制造新質(zhì)控圖。根據(jù)滴度或稀釋度斷定陰陽性的技術(shù)，如凝集實驗，每檢測日或分析批，應(yīng)運用弱陽性和陰

5、性外對照作為質(zhì)控。實驗室應(yīng)定義本人的質(zhì)控批長度。陽性質(zhì)控結(jié)果在均值上下一個滴度或稀釋度以及陰性質(zhì)控結(jié)果為陰性即為在控，否那么視為失控。純定性實驗如金標(biāo)試紙、斑點滲濾等，檢測安裝的內(nèi)對照外，每檢測日或分析批，應(yīng)運用弱陽性和陰性外對照作為質(zhì)控。實驗室應(yīng)定義本人的質(zhì)控批長度。陰、陽性質(zhì)控的檢測結(jié)果分別為陰性和陽性即闡明在控，相反那么為失控。應(yīng)評價失控形狀下病人結(jié)果的影響。臨床血液學(xué)5.6.1室內(nèi)質(zhì)量控制應(yīng)符合如下要求：質(zhì)控品的選擇：引薦運用配套質(zhì)控品，運用非配套質(zhì)控品時應(yīng)評價其質(zhì)量和適用性；質(zhì)控品的濃度程度：至少運用2個濃度程度正常和異常程度的質(zhì)控品；質(zhì)控工程：申報認(rèn)可的一切檢測工程均應(yīng)開展室內(nèi)質(zhì)量

6、控制；質(zhì)控頻度：應(yīng)經(jīng)過檢測質(zhì)控物，檢查血常規(guī)檢驗的精細(xì)性，應(yīng)根據(jù)實驗室檢驗標(biāo)本的數(shù)量定期實施，檢測當(dāng)天至少1次；質(zhì)控方法：應(yīng)運用Levey-Jennings質(zhì)控圖；Levey-Jennings質(zhì)控圖或類似的質(zhì)量控制記錄應(yīng)包含以下信息：檢測質(zhì)控品的時間范圍、質(zhì)控圖的中心線和控制界限、儀器/方法稱號、質(zhì)控品的稱號、濃度程度、批號和有效期、試劑稱號和批號、每個數(shù)據(jù)點的日期、操作人員的記錄。質(zhì)控圖中心線確實定：血常規(guī)檢查的質(zhì)控品測定應(yīng)在每天的不同時段至少檢測3天，至少運用10個檢測結(jié)果的均值為質(zhì)控圖的中心線；凝固實驗的質(zhì)控品至少檢測10天，至少運用20個檢測結(jié)果的均值為質(zhì)控圖的中心線；凝固實驗改換新批

7、號試劑或儀器進展重要調(diào)整時，應(yīng)重新確定質(zhì)控品的均值；每個新批號的質(zhì)控品在日常運用前，應(yīng)由實驗室經(jīng)過檢測確定質(zhì)控品均值，制造商規(guī)定的“規(guī)范值只能作為參考，通常實驗室確定的質(zhì)控品均值宜在配套定值質(zhì)控品的允許范圍內(nèi)。質(zhì)控品均值的計算方法參見GB/T20468-2006。規(guī)范差確實定：經(jīng)過一段時間的反復(fù)檢測確定室內(nèi)質(zhì)量控制的規(guī)范差，規(guī)范差的計算方法參見GB/T20468-2006。失控的判別規(guī)那么：實驗室應(yīng)有程序規(guī)定運用的質(zhì)控規(guī)那么，至少運用13s 規(guī)那么22s，多種質(zhì)量控制規(guī)那么的運用可以提高誤差檢出概率。失控報告應(yīng)包括失控情況的描畫、核對方法、緣由分析、糾正措施及糾正效果的評價等內(nèi)容。質(zhì)控數(shù)據(jù)的管

8、理：原那么上每月統(tǒng)計1次，至少保管二年。記錄的檢查：血液學(xué)檢驗部門的擔(dān)任人或由擔(dān)任人指定的授權(quán)人應(yīng)至少每月對室內(nèi)質(zhì)量控制的記錄進展檢查并簽字。臨床體液學(xué)5.6.1 尿有構(gòu)成份分析儀紅細(xì)胞、白細(xì)胞計數(shù)的室內(nèi)質(zhì)控，可參照GB/T 20468 -2006。應(yīng)至少運用2 個濃度程度正常和異常程度的質(zhì)控品，每任務(wù)日至少1次。實驗室應(yīng)至少運用13s、22s失控規(guī)那么。對于定性體液學(xué)檢測工程的室內(nèi)質(zhì)控，應(yīng)至少運用陰性和陽性質(zhì)控品，檢測當(dāng)天至少1次，偏向不超越1個等級，且陰性不可為陽性，陽性不可為陰性。臨床微生物學(xué)5.6.1 實驗室制定的內(nèi)部質(zhì)量控制程序應(yīng)包括整個實驗操作過程，照實驗分析前、中、后報告。細(xì)菌的

9、質(zhì)控應(yīng)滿足如下要求：運用中的染色劑革蘭染色,特殊染色，熒光染色，至少每周用知陽性和陰性適用時的質(zhì)控菌株檢測染色程序。凝固酶、過氧化氫酶、氧化酶、內(nèi)酰胺酶運用時，每天應(yīng)做陰性和陽性質(zhì)控，商業(yè)頭孢菌素試劑的內(nèi)酰胺酶實驗可遵照制造商的建議。診斷性抗血清實驗應(yīng)設(shè)陰性、陽性對照。實驗室常規(guī)采用的藥敏實驗方法應(yīng)以參考菌株延續(xù)檢測2030天，每一組藥物/微生物超出參考范圍抑菌圈直徑或MIC的頻率需小于1/20或2/30。以后，應(yīng)每周運用規(guī)范菌株對藥敏實驗進展質(zhì)控。應(yīng)保管抗菌藥物敏感性實驗資料，并至少每年向臨床醫(yī)師報告。厭氧菌應(yīng)以有效的方法檢測厭氧培育環(huán)境如以亞甲蘭試條、厭氧菌或適當(dāng)?shù)某绦驒z測厭氧系統(tǒng)的厭氧條

10、件。分枝桿菌抗酸染色應(yīng)在實驗的每天用適當(dāng)?shù)年栃院完幮再|(zhì)控驗證；熒光染色應(yīng)每次以陰性和陽性對照驗證。真菌直接染色如：抗酸染色，PAS，吉姆薩染色，墨汁染色檢查患者標(biāo)本時，應(yīng)每天做陰性和陽性質(zhì)控某些染色如吉姆薩染色，玻片本身作為陰性質(zhì)控。KOH制備的玻片不需求質(zhì)控。病毒延續(xù)細(xì)胞傳代時應(yīng)定期監(jiān)測支原體污染宜監(jiān)測陰性未傳代的質(zhì)控株，而不是培育支原體；應(yīng)監(jiān)測用于細(xì)胞生長培育液的動物血清的細(xì)胞毒性；應(yīng)具備相應(yīng)的細(xì)胞株用于病毒培育?；驍U增檢驗實驗室應(yīng)制定室內(nèi)質(zhì)量控制程序，可參照GB/T 20468 -2006。質(zhì)控圖應(yīng)包括質(zhì)控結(jié)果、質(zhì)控品稱號、濃度、批號和有效期、質(zhì)控圖的中心線和控制界限、分析儀器稱號和獨

11、一標(biāo)識、方法學(xué)稱號、檢驗工程稱號、試劑和校準(zhǔn)品批號、每個數(shù)據(jù)點的日期和時間、干擾行為的記錄、質(zhì)控人員及審核人員的簽字、失控時的分析處置程序和糾正措施等。對于定性檢測工程，每次實驗應(yīng)設(shè)置陰性、弱陽性和陽性質(zhì)控。質(zhì)控規(guī)那么應(yīng)確保實驗的穩(wěn)定性和檢驗結(jié)果的可靠性。實驗室應(yīng)檢查失控對之前檢驗的影響。統(tǒng)計質(zhì)量控制控制品控制圖均值與規(guī)范差控制規(guī)那么質(zhì)控頻率失控后的處置控制品正確度控制品正確度控制品Trueness Control不同于普通的控制品。在定值要求上等同于校準(zhǔn)品的定值。公用于核實、確認(rèn)、證明檢測系統(tǒng)能否真實地實現(xiàn)了溯源性。公用性廠商能否有才干提供“正確控制品，是衡量廠商提供的檢測系統(tǒng)產(chǎn)品能否可以使

12、實驗室每天的患者標(biāo)本檢驗結(jié)果具有溯源性的重要標(biāo)志。Trueness (measurement) closeness of agreement between the average of an infinite number of replicate measured quantity values and a reference quantity value(JCGM 200:2021); NOTE:Trueness is usually expressed numerically by the statistical bias, which is inversely related to

13、trueness.trueness control material a reference material with measurand value assigned by a procedure traceable to a higher order reference system and with commutability properties suitable for use to assess the bias of measurement of a specified measurement procedure(modified from ISO 17511).Truenes

14、s Controls Target values may also be assigned to reference (or QC sample)materials by reference measurement laboratories using certified primary reference measurement procedures . These control samples may be used to verify the trueness of a laboratorys measuring system, as long as their commutabili

15、ty with patient samples has been validated for a given measurement procedure. Trueness controls should also have product labeling that starts the materials are intended for trueness of measurement, and identifies the measuring systems for which the commutability was validated.Trueness ControlsTruene

16、ss controls are generally too expensive for routine use in QC, but they are invaluable for verifying that a measuring system is properly calibrated when it is first implemented or periodically thereafter. Trueness controls are also useful for routine verification of calibration or troubleshooting wh

17、en the accuracy of patient results is suspect. 控制品定值控制血清定量控制品是在曾經(jīng)確認(rèn)了多家穩(wěn)定的檢測系統(tǒng)有的可溯源的前提下對控制品進展定值。定量控制品只需在一樣穩(wěn)定檢測系統(tǒng)上運用才具有估計結(jié)果可靠性的參考價值，也僅僅是參考價值。在其它檢測系統(tǒng)上，這個定值沒有參考價值！更不可運用定值控制品為校準(zhǔn)品！Control Materials with Assigned ValuesControl Materials with Assigned Values (Assayed Quality Control Samples)Measuring system

18、 specific values can also be assigned to QC samples as target values. such materials are often called “assayed controls , and the values are assigned by the QC material manufacturer or by laboratories in a value assignment program using a given measuring system . These material are intended to give

19、individual laboratories a means of determining whether their performance is as expected for a given measuring system. The usefulness of these system-specific values depends on the traceability and uncertainty of the assigned values.復(fù)溶過程中多種要素對干粉質(zhì)控品檢測結(jié)果的影響評價資料干粉質(zhì)控品：邁瑞正常值批號QMN295和病理值干粉質(zhì)控品(批號QMP296)，規(guī)格為

20、5/ml瓶。儀器：BS-800全自動生化分析儀深圳邁瑞公司，Microlab 500稀釋配液儀美國HAMILTON公司，QB208型自動混勻器(海門其林倍爾公司)。檢測試劑：ALB，ALP，ALT，AMY,AST, Ca，CK，Cl，CREA，DBIL， Glu， K，LDH，Mg，Na，P，TBIL，TC，TG，TP，UA, UREA， HBDH，GGT為邁瑞公司原裝試劑。復(fù)溶過程中多種要素對干粉質(zhì)控品檢測結(jié)果的影響評價結(jié)論實驗人員復(fù)溶操作過程對干粉質(zhì)控品的檢測結(jié)果有影響復(fù)溶時的溫度以室溫242水質(zhì)選用電導(dǎo)率1 s/cm的純水為宜控制圖控制圖L-J控制圖Z值圖Youden圖均值和規(guī)范差的設(shè)置

21、均值和規(guī)范差的統(tǒng)計固定均值和規(guī)范差固定均值和SD:每月恒定的均值3SD范圍設(shè)定質(zhì)控圖的均值和控制限各實驗室應(yīng)運用本室現(xiàn)行的檢測系統(tǒng)，對新批號的控制品的各個測定工程自行確定均值和控制限。控制限通常用規(guī)范差的倍數(shù)來表示，臨床實驗室不同定量測定工程的控制限要根據(jù)其采用的控制規(guī)那么來決議。設(shè)定質(zhì)控圖的均值和控制限 穩(wěn)定性較長的控制品做法1第三版全國臨床檢驗操作規(guī)程：對新批號的控制品進展測定，根據(jù)20或更多批獲得的質(zhì)控測定結(jié)果，進展離群值檢驗剔除超越3s外的數(shù)據(jù)，計算出平均數(shù)和規(guī)范差，作為暫定均值和暫定規(guī)范差。以此暫定均值和規(guī)范差作為下一月室內(nèi)質(zhì)控圖的均值和規(guī)范差進展室內(nèi)質(zhì)控。設(shè)定質(zhì)控圖的均值和控制限穩(wěn)

22、定性較長的控制品做法1第三版全國臨床檢驗操作規(guī)程：一個月終了后，將該月的在控結(jié)果與前20個質(zhì)控測定結(jié)果聚集在一同，計算累積平均數(shù)和累積規(guī)范差第一個月，以此累積的平均數(shù)和規(guī)范差作為下一個月質(zhì)控圖的均值和規(guī)范差。反復(fù)上述操作過程，延續(xù)3至5個月，以最初20個數(shù)據(jù)和3至5個月在控數(shù)據(jù)聚集的一切數(shù)據(jù)計算累積平均數(shù)和規(guī)范差，以此作為該控制品有效期內(nèi)的常用均值和常用規(guī)范差。設(shè)定質(zhì)控圖的均值和控制限 穩(wěn)定性較長的控制品做法2：對新批號的控制品進展測定，根據(jù)20或更多批獲得的質(zhì)控測定結(jié)果，進展離群值檢驗剔除超越3s外的數(shù)據(jù)，計算出平均數(shù)和規(guī)范差，作為暫定均值和暫定規(guī)范差。以此暫定均值和規(guī)范差作為下一月室內(nèi)質(zhì)控

23、圖的均值和規(guī)范差進展室內(nèi)質(zhì)控。設(shè)定質(zhì)控圖的均值和控制限穩(wěn)定性較長的控制品做法2：一個月終了后，將該月剔除人為錯誤數(shù)據(jù)、超越3s外數(shù)據(jù)后一切結(jié)果與前20個質(zhì)控測定結(jié)果聚集在一同，計算累積平均數(shù)和累積規(guī)范差第一個月，以此累積的平均數(shù)和規(guī)范差作為下一個月質(zhì)控圖的均值和規(guī)范差。反復(fù)上述操作過程，延續(xù)3至5個月，以最初20個數(shù)據(jù)和3至5個月剔除人為錯誤數(shù)據(jù)、超越3s外數(shù)據(jù)后一切數(shù)據(jù)聚集，計算累積平均數(shù)和規(guī)范差，以此作為該控制品有效期內(nèi)的常用均值和常用規(guī)范差。設(shè)定質(zhì)控圖的均值和控制限 穩(wěn)定性較長的控制品做法3：對新批號的控制品進展測定，根據(jù)20或更多批獲得的質(zhì)控測定結(jié)果，進展離群值檢驗剔除超越3s外的數(shù)據(jù)

24、，計算出平均數(shù)和規(guī)范差，作為暫定均值和暫定規(guī)范差。以此暫定均值和規(guī)范差作為下一月室內(nèi)質(zhì)控圖的均值和規(guī)范差進展室內(nèi)質(zhì)控。設(shè)定質(zhì)控圖的均值和控制限穩(wěn)定性較長的控制品做法3：一個月終了后：將該月剔除人為錯誤數(shù)據(jù)、超越3s外數(shù)據(jù)后一切結(jié)果與前20個質(zhì)控測定結(jié)果聚集在一同，計算累積平均數(shù)X1和累積規(guī)范差(S1)第一個月；將該月的在控結(jié)果與前20個質(zhì)控測定結(jié)果聚集在一同，計算累積平均數(shù)X2和累積規(guī)范差(S2) 第一個月；以X2和S1作為下一個月質(zhì)控圖的均值和規(guī)范差。反復(fù)上述操作過程，延續(xù)3至5個月：以最初20個數(shù)據(jù)和3至5個月剔除人為錯誤數(shù)據(jù)、超越3s外數(shù)據(jù)后一切數(shù)據(jù)聚集，計算累積平均數(shù)X3和規(guī)范差S3;

25、以最初20個數(shù)據(jù)和3至5個月在控數(shù)據(jù)聚集的一切數(shù)據(jù)計算累積平均數(shù)X4和規(guī)范差S4；以X4和S3作為該控制品有效期內(nèi)的常用均值和常用規(guī)范差。設(shè)定質(zhì)控圖的均值和控制限能否必需對“定值控制品建立均值和范圍？CLIA 只需靶值，只適用于運用的方法和儀器。必需經(jīng)過反復(fù)檢測確定均值與規(guī)范差。CLSI 定值僅供參考，必需經(jīng)過反復(fù)檢測建立本人的值。臨床實驗室定量測定室內(nèi)質(zhì)量控制指南 假設(shè)運用定值質(zhì)控品，運用闡明書上的原有標(biāo)定值只能作參考。必需由實驗室作反復(fù)測定來確定實踐的均值和規(guī)范差。設(shè)定質(zhì)控圖的均值和控制限 穩(wěn)定性較短的控制品在至少3至4天內(nèi)，每天分析每程度控制品3至4瓶，每瓶進展2至3次反復(fù)。搜集數(shù)據(jù)后，

26、計算平均數(shù)、規(guī)范差和變異系數(shù)。對數(shù)據(jù)進展離群值檢驗剔除超越3S的數(shù)據(jù)。假設(shè)發(fā)現(xiàn)離群值，需重新計算余下數(shù)據(jù)的平均數(shù)和規(guī)范差。以此均值作為質(zhì)控圖的均值。對于穩(wěn)定性較短的控制品，應(yīng)采用以前變異系數(shù)CV%來估計新的規(guī)范差，即規(guī)范差等于平均數(shù)乘以變異系數(shù)CV%。 改換控制物改換新批號控制物時，應(yīng)在“舊批號控制物運用終了前，將新批號控制物與“舊批號控制物同時進展測定。反復(fù)上面過程，設(shè)立新的X、SD。改換質(zhì)控品，對新批號的控制品進展測定，根據(jù)20或更多批獲得的質(zhì)控測定結(jié)果。改換質(zhì)控品，假設(shè)無法從20天內(nèi)得到20個數(shù)值，至少在5天內(nèi)，每天作不少于4次反復(fù)檢測來獲得。即刻性質(zhì)控法 對于某些不是每天開展的工程或試

27、劑盒有效期較短的工程可采用即刻性質(zhì)控方法，只需延續(xù)測定3次，即可對第三次及以后的檢驗結(jié)果進展控制。即刻性質(zhì)控法評價適用性：正態(tài)分布或近似正態(tài)分布，檢測頻次低的定量檢驗工程或定性實驗缺陷：結(jié)果易受前3個質(zhì)控數(shù)據(jù)的影響，在統(tǒng)計學(xué)中樣本量少時容易出現(xiàn)抽樣誤差，Grubbs檢驗法要求樣本數(shù)據(jù)不可少于6個?；叵胄允Э貦z出：遲后異常值就是某一測定值當(dāng)天用Grubbs檢驗法判別為極值，在控，但隨著數(shù)據(jù)的添加，后來判別為異常值，屬于“警告或“失控。浮動均值和浮動規(guī)范差的設(shè)置浮動均值和規(guī)范差浮動均值和SD:恒定Z=3ALB、Na+、C1-、Ca CREA、URIC、Na+、 C1- GLU、TBIL、CREA、

28、URIC ALB FSH2程度浮動變異系數(shù)CV與值時間浮動cv偏倚值SE性能質(zhì)控規(guī)則NPedPfr1月31日3.566.195.283.63良好13S,22S,R4S,41S,10X210.0132月15日4.446.194.282.63良好13S,22S,R4S,41S,10X20.920.0122月28日4.396.194.282.63良好13S,22S,R4S,41S,10X20.9910.0133月15日4.736.193.982.33良好13S,22S,R4S,41S,10X20.9070.0133月31日5.066.193.722.07良好13S,22S,R4S,41S,10X20

29、.9150.0134月15日5.876.193.21.55良好13S,22S,R4S,41S,10X20.6770.0134月30日6.266.1931.35良好13S,22S,R4S,41S,10X20.5280.0135月15日6.26.193.131.38良好13S,22S,R4S,41S,10X20.5280.0135月31日6.226.193.021.37良好13S,22S,R4S,41S,10X20.5280.0136月15日6.276.1931.35良好13S,22S,R4S,41S,10X20.5280.0136月30日6.496.192.91.25合格13S,22S,R4S,

30、41S,10X20.450.0137月15日6.656.192.831.18差13S,22S,R4S,41S,10X20.3720.0137月31日6.686.192.821.17差13S,22S,R4S,41S,10X20.3720.0138月15日6.616.192.851.2差13S,22S,R4S,41S,10X20.3720.0138月31日6.46.192.941.29合格13S,22S,R4S,41S,10X20.5060.01915個化學(xué)發(fā)光工程8個月每月監(jiān)測兩次質(zhì)控結(jié)果浮動CV評價21個常規(guī)生化工程延續(xù)20次質(zhì)控結(jié)果浮動CV%評價 C24-A3 Statistical Qua

31、lity Control for Quantitative Measurements8.6.5 Cumulative Values Estimates of the standard deviation (and to a lesser extent the mean) from monthly control data are often subject to considerable variation from month to month, due to an insufficient number of measurements(e.g., with 20 measurements,

32、 the estimate of the standard deviation might vary up to 30% from the true standard deviation;even with 100 measurements, the estimate may vary by as much as 10%). More representative estimates can be obtained by calculating cumulative values based on control data from longer periods of time (e.g.,

33、combining control data from a consecutive six-month period to provide a cumulative estimate of the standard deviation of the measurement procedure). This cumulative value will provide a more robust representation of the effects of factors such as recalibration, reagent lot change,calibrator lot chan

34、ge, maintenance cycles,and environmental factors including temperature and humidity.Care should be taken to ensure that the method has been stable and the mean is not drifting consistently lower or consistently higher over the six-month periods being combined, for example due to degradation of the c

35、alibrator or control material.浮動均值和浮動規(guī)范差控制圖控制圖L-J控制圖(固定/浮動)Z值圖(固定/浮動)Youden圖(固定/浮動)控制規(guī)那么Westgard多規(guī)那么質(zhì)控方法修正的Westgard多規(guī)那么質(zhì)控方法SPC規(guī)那么Westgard多規(guī)那么質(zhì)控方法修正的Westgard多規(guī)那么質(zhì)控方法測定次數(shù)和控制規(guī)那么的選擇操作過程規(guī)范圖(OPSpecs圖):測定次數(shù)N和控制規(guī)那么的選擇應(yīng)盡量采用測定次數(shù)N=2或N=3,能檢出90%的系統(tǒng)誤差,同時維持盡能夠5%的假失控率。對于方法性能“差的工程，有必要重新進展方法學(xué)評價，重新選擇分析方法或不同廠家試劑盒。對于測定方

36、法的偏向過大，導(dǎo)致達不到臨床質(zhì)量要求。OPSpecs圖OPSpecs圖TEa的設(shè)置：CLIA，生物學(xué)變異，自定CV的設(shè)置：累計CVBias%設(shè)置相近濃度的室間質(zhì)評Bias%控制品室間比對的對等組或一樣方法組的Bias%室間比對的Bias%當(dāng)前檢測系統(tǒng)測定互通性好的參考物質(zhì)與靶值的Bias%新穎標(biāo)本當(dāng)前檢測系統(tǒng)測定值與參考方法賦值的Bias%例如(Alb實踐CV和BIAS)例如Alb實踐CV,BIAS(1/3TEa)例如Alb實踐CV,BIAS(1/2TEa)例如AlbCV(1/4TEa),BIAS(1/3TEa)例如Alb-CV(1/3TEa),BIAS(1/2TEa)例如AlbCV(1/3T

37、Ea),BIAS(1/3TEa)例如AlbCV(1/4TEa),BIAS(1/2TEa)質(zhì)控頻率-C24-A3 Frequency of controlmeasurementsQuality control samples must be analyzed at least once during each user-defined analytical run length. Manufacturers of analytical systems or reagents may recommend the number of quality specimens and their locat

38、ion within the run. However,manufacturers recommendations should be used as guidelines and frequency of QC measurement should be established by the laboratory considering the factors outlined in section 7.2. The frequency and location of control samples should reflect actual test system performance

39、and application at the site of testing. The user may need additonal control specimens and a different location in order to meet different laboratory circumstances. 質(zhì)控頻率-EP23-A lab Qulity Control Based on Risk Management 5.1.6 Frequency of Quality Control Sample TestingThe optimal frequency of perfor

40、ming QC sample procedures depends on built-in and other controls for a given measuring system, the stability of that measuring system, and conditions in the laboratory identified through risk assessment that could affect the reliability of testing such as staff turnover, and the clinical risk of har

41、m to a patient if an erroneous result is reported and acted on. The frequency of control procedures should also conform to applicable regulatory and accreditatlbn requirements. Monitoring the measuring system at shorter intervals increases the likelihood that systematic errors are detected before in

42、correct results are reported, or decreases the time before alerting the health care provider who may have received incorrect results. For example, a laboratory that evaluates the examination (analytical) process every eight hours will identity a systematic error condition much earlier than a laborat

43、ory that monitors the examination process every 24 hours. However, the total number of specimens tested in a time interval may also influence the frequency of monitoring. For example, a laboratory that tests 2000 samples in one 24-hour period might perform control procedures several times a day, whe

44、reas a laboratory that tests 50 samples in an eight-hour shift might perform control procedures at the beginning and end of a shift. The complex relationship between the frequency of QC sample testing, frequency of false rejection,and the quality of patient results has been explored by Parvin and co

45、lleagues.EP23-A:6.3.3 Frequency of Patient TestingThe QCP should take into account the frequency with which patient testing is performed. Operators may be less familiar with the details of a procedure performed infrequently, thereby increasing the probability of use errors and erroneous result. Freq

46、uent repetition of the steps of a measuring system leads to operator competency. Consequently, when a test is not performed often in a laboratory, consider the need for more frequent QC sample testing. Give consideration to the simplicity/complexity of the procedure and whether the procedure is full

47、y automated, semiautomated, or manual. Dilutions, extractions, calculations, and maintenance of measuring systems and support equipment(refrigerators, centrifuges, balances), for example, can be challenging to operators. Manual steps typically pose more risks than automated processes owing to the li

48、kelihood of use errors. 我院的質(zhì)控頻率與控制檔次置分析批：一段時間的區(qū)間，或是一組病人標(biāo)本量的大小，統(tǒng)計過程確定控制形狀的對象。CLIA88規(guī)定，臨床化學(xué)檢驗最大批的時間為24小時，血液學(xué)檢驗為8小時。我科的做法：門診PPI常規(guī)化學(xué)每天3次，上午7：30左右兩程度Biorad質(zhì)控物，與標(biāo)本同時進展；中午10:30 11:00左右做1個程度正常值配套質(zhì)控物；下午4時左右儀器保養(yǎng)終了，加試劑后做1個程度病理值配套質(zhì)控物。門診化學(xué)法光Centaur每天3次，上午7：30一程度Biorad正常值質(zhì)控物，與標(biāo)本同時進展；中午10:30 11:00左右大部分標(biāo)本快終了時做1個程度病

49、理值Biorad質(zhì)控物；下午4時左右儀器保養(yǎng)終了，加試劑后做1個程度正常值Biorad質(zhì)控物。門診化學(xué)法光601每天2次，上午7：30做1個程度正常值Biorad質(zhì)控物，與標(biāo)本同時進展；中午11:30左右大部分標(biāo)本快終了時做1個程度病理值Biorad質(zhì)控物。我院的質(zhì)控頻率與控制檔次置病房PPI常規(guī)化學(xué)每天3次，上午7：30左右兩程度配套質(zhì)控物，在控后檢測標(biāo)本；中午10:3011:00左右做兩個程度Biorad質(zhì)控物。病房干化學(xué)上午8時左右做兩程度配套質(zhì)控物。下午2：30左右做兩程度Biorad質(zhì)控物，24小時檢測標(biāo)本。病房特種蛋白每天一次，每次兩個程度配套質(zhì)控物，在控后檢測標(biāo)本。病房糖化血紅蛋

50、白每天一次，每次兩個程度，兩程度配套質(zhì)控物或兩程度Biorad質(zhì)控物交替進展，在控后檢測標(biāo)本。病房化學(xué)法光上午8時左右做兩程度Biorad質(zhì)控物，24小時檢測標(biāo)本。病房血氣分析每天1次，每次一個程度，配套質(zhì)控物和Biorad質(zhì)控物，交替進展，24小時檢測標(biāo)本。失控后的處置失控后的不當(dāng)做法預(yù)防性質(zhì)控失控緣由分析失控后的處置分析目的質(zhì)量控制提高檢驗質(zhì)量統(tǒng)計質(zhì)量控制方法分析目的質(zhì)量控制方法累積均值和規(guī)范差浮動均值和規(guī)范差不精細(xì)度-生物學(xué)變異總誤差-生物學(xué)變異當(dāng)前技術(shù)程度醫(yī)學(xué)相關(guān)性常用研討很少少用分析目的質(zhì)量控制當(dāng)前技術(shù)程度：根據(jù)當(dāng)前可以到達的檢測不精細(xì)度，將測試不精細(xì)度控制在目的范圍內(nèi)。 不精細(xì)度-生物學(xué)變異：根據(jù)生物學(xué)變異數(shù)據(jù)制定實驗室本人的性能目的。 總誤差-生物學(xué)變異：根據(jù)由生物學(xué)變異確定的總允許誤差制定檢測性能的上下控制限。醫(yī)學(xué)相關(guān)性：根據(jù)臨床規(guī)范制定分析過程的總允許誤差。CV很小，假失控率太高，浪費人力物力。如TE =10%，CV = 1%統(tǒng)計質(zhì)量控制存在的問題CV很大，失控檢出率會很低。如TE=10%， CV=7%統(tǒng)計質(zhì)量控制存在的