《质量控制概要.ppt》由会员分享，可在线阅读，更多相关《质量控制概要.ppt（73页珍藏版）》请在三一文库上搜索。
1、1,Review of Quality Control质量控制概要,We will cover(内容) Terminology 术语 Characteristics of Analytical Methods 分析方法特性 Types of QC samples and how they relate to the characteristics QC样品类型和特性,天马行空官方博客： ;QQ:1318241189,2,Definition of Quality质量的定义,To most of the world (普遍上) degree of excellence(优异程度) To the
2、ISO world (在ISO领域) degree of agreement between characteristics and requirements 特性与要求的吻合程度 More importantly (更重要) to our clients?(对客户),3,Quality Control vs. Quality Assurance(质量控制与质量保证),Quality control is a series of measurements used to assess the quality of the analytical data 质量控制是一系列测量,以评价分析数据的质
3、量 How well is the test fulfilling requirements? 试验满足要求的程度如何?,4,Quality Control vs. Quality Assurance (质量控制与质量保证),Quality assurance is a management plan to ensure the integrity of data (a system) 质量保证是一种管理计划,以确保数据的完整性(一种体系) Provides confidence that requirements will be fulfilled 为要求是否得到满足提供信心,天马行空官方博
4、客： ;QQ:1318241189,5,Qualities of Analytical Methods分析方法的质量,Accuracy(准确度) Precision(精密度) Selectivity(选择性) Sensitivity(灵敏度) Linearity(线性) Range(范围) Robustness / Ruggedness(稳健性/稳度),6,Accuracy(准确度),Degree of agreement between the “true” value and the measured value(真值与测量值之间的一致程度),天马行空官方博客： ;QQ:131824118
5、9,7,Accuracy (准确度),What is the “true” value?(什么是真值) Theoretical(理论值) Consensus(公议值) Accepted / Assigned(可接受/赋值),8,Accuracy (准确度),Expressed as (表示为) the ratio between the assayed value and the “true” value (% recovery) 分析值与真值间的比(%回收率) the difference between the assay and the accepted true value (% Er
6、ror) 分析与可接受真值间的差(%误差),9,Precision(精密度),Closeness of agreement between measurements(测量间的一致程度) Repeatability (r) same conditions, short time interval (same run) short term precision 重复性(r)-相同条件下短时间间隔-短期内精密度 Intermediate precision same lab different day, analyst, instrument 中间精密度-同一实验室不同日期,分析人员,仪器 Repr
7、oducibility (R) agreement between batches or between labs 复现性-批间或实验室间的一致性,天马行空官方博客： ;QQ:1318241189,10,Precision (精密度),Expressed as Standard Deviation, Relative Standard Deviation or Relative Percent Difference 表示为标准偏差、相对标准偏差、相对百分差,11,Sensitivity（灵敏度）,Degree of response to a change in analyte concent
8、ration resolution 被分析物浓度变化的响应变化程度-分辩率 Lowest concentration that can be distinguished from background noise detection limit 能从背景噪音识别出的最低浓度-检测限,12,Sensitivity （灵敏度）,Limit of detection vs. Limit of quantification（检测限与定量限） Limit of detection (LOD) is the level that we can say whether an analyte is prese
9、nt or not 检测限是我们判断被分析物是否存在的含量水平 Limit of quantitation (LOQ) is the level that we can quantify the analyte with a specified accuracy (20%) 定量限是我们按规定准确度（20%）能对被分析物量化的含量水平,13,Selectivity（选择性）,Ability to detect the analyte in a complex matrix 检测复杂基体中被析物的能力 Ability to exclude non-analyte species from sig
10、nal 通过信号排除非被分析物的能力 Usually inversely related to sensitivity 通常与灵敏度成反比,14,Linearity（线性）,Ability to produce a response that is directly proportional to the concentration of the analyte 所产生的反应与被分析物浓度成正比的能力 Only holds true within a given range 只在给定范围内是真实的,15,Linearity （线性）,Expressed as correlation coeff
11、icient (r) or coefficient of determination (r2) 表示为相关系数(r)或测定系数(r2),r2=ssxy2/ssxx ssyy,16,Range（范围）,The interval between the highest and lowest concentration for which the method has shown to have suitable levels of accuracy, precision and linearity. 在方法具有适宜的准确度、精密度和线性的条件下，最高和最低浓度范围,17,Robustness（稳健
12、性）,Ability of method to be unaffected by small changes in method (internal) conditions（方法不因条件的较小变化（内部）而受影响的能力） Temperature（温度） pH Composition of mobile phase（流动相组成） Useful when method moves from development to production（从方法开发到使用很有用）,18,Ruggedness（稳度）,Ability of method to be unaffected by external c
13、onditions（方法不受外界条件影响的能力） Different analysts（不同的分析人员） Different laboratories（不同的实验室） Different reagent lots（不同批的试剂） Different instruments（不同的仪器）,19,Ruggedness （稳度）,Useful for technology transfer（对技术推广很有用）,20,Quality Control Measures质量控制手段,Calibration standards（校准标准） Blanks（空白） Recovery studies（回收率研究）
14、 Precision studies（精密度研究） Sensitivity studies（灵敏度研究） Detection Limit（检测限） Quantitation limit（定量限）,21,Calibration (校准),Directly affects accuracy(直接影响准确度) Preparation of standards(标准制备) Matrix (基体) Stability(稳定性) Signal type: peak area, peak height 信号类型:峰面积,峰高,22,Calibration (校准),Calibration curves(校准
15、曲线) 1st order straight line(一级-直线) Quadratic acceptable but requires more points(二次方程-可以接受,但需要更多点) Internal standards(内标) Refines calibration response(增强校准响应) Added to samples and standards(加入样品和标准中) Must behave like the analyte(象被分析物一样表现) Cannot be in samples rare elements, deuterated compounds(样品中
16、没有-稀有元素,含重氘的化合物),23,Calibration (校准),Calibration verification (校准验证) Repeated calibration standard 重复校准标准 Interval may be every 20 samples (ICPMS) or once every run (GC-FID) 间隔:20个样品(ICPMS) 每次(GC-FID),24,Blanks(空白),Used to monitor contamination(监测污染) Reagent blanks(试剂空白) Instrument blanks(仪器空白) Fiel
17、d blanks(场地空白) Trip blanks(路途空白) Procedural / method blanks(程序/方法空白) Matrix blanks(基体空白) Others?(其它),25,Recovery Studies（回收率研究）,Used to monitor accuracy（监测准确度） Matrix spikes（基体加标） Fortified blanks（增加加标） Surrogates（替代） Reference Materials（参考物质）,26,Matrix Spikes（基体加标）,Known amount of analyte added to
18、sample 在样品中加入已知量的被分析物 Important with complex matrix 对复杂基体很重要 Generally added prior to extraction process（通常在提取前加入） Sometimes difficult to know amount to spike ahead of time（有时很难提前知道加标量）,27,Fortified Blanks（增强空白）,Analyte spike into reagent blank 在试剂空白中加入被分析物标样 Carried through sample preparation proce
19、ss 在样品准备过程中加入,28,Surrogate（替代）,Compound with similar properties to target analyte 与被分析物具有相近特性的化合物 Added to all samples (unknowns and QC) but not standards 在所有样品(未知的和QC)中加入，但标样中不加 Used to monitor method performance on a sample (matrix) specific basis 应基于特定样品（基体），用于监测方法表现,29,Reference Materials参考物质,Ma
20、trix matched materials with known concentration of analytes（被分析物已知浓度的基体匹配的物质） May be external or in-house（外部 或室内） Sources（来源） NIST (US) NRC (Canada) European Union（欧盟） Commercial Vendors（商业渠道）,30,Reference Materials参考物质,Used primarily to monitor accuracy 主要用于监测准确度 May be used for other quality 用于其它质
21、量目的 measures (precision, sensitivity) 测量（精密度、灵敏度） Best overall QC?（最好的全面质量控制？） Expensive（费用高）,31,Precision Studies（精密度研究）,Duplicates（重复） Field duplicates（场地重复） Lab duplicates（实验室重复） Instrument duplicates（仪器重复） Inter-lab splits（室内分样） Used to identify variance in different processes（用于识别不同过程的变异）,32,Se
22、nsitivity Studies灵敏度研究,Usually performed initially, then periodically thereafter 通常最开始要做，以后定期做 Multiple replicates of a blank or low-level standard 空白或低含量标准多次重复,33,Sensitivity Studies灵敏度研究,Detection Limit no quantitation 检测限-非定量 Quantitation Limit Accuracy criteria added 定量限准确度准则,34,Sensitivity Stud
23、ies灵敏度研究,Detection Limit 99% certainty that the concentration is greater than zero that the analyte is present in the sample（检测限99%确定浓度大于0-样品中有被分析物） Uncertainty is high (+/- 100% or more) 不确定度高（ +/- 100% 或更高） Quantitation Limit（定量限） Lowest concentration that can be determined with a given accuracy (
24、e.g. 20% relative error) 在一定准确度下（如20%相对误差）可以测定出的最低浓度,35,Linearity（线性）,Determined initially and then as needed 最开始时测定，然后需要时测定 Dynamic Linear Range relates to detector response 动态线性范围与检测器响应有关 Also consider preparation limitations 也可以认为是准备中的限制 Solubilities（溶解度） Extraction efficiencies（提取率） Derivitizing
25、 reactions（衍生反应）,36,What do we run as QC?为什么要做质量控制,It depends on:（取决于） Type of instrumentation 仪器类型 Stage of Method Development 方法开发阶段 Type of analysis 分析类型,37,What do we run as QC?为什么要做质量控制,Stage of Method Development 方法开发阶段 Validation（确认） Verification / Qualification（验证/判断） Routine analysis（日常分析）,
26、38,What do we run as QC?为什么要做质量控制,Type of Analysis（分析类型） Assay（分析） Impurities（杂质） Limits（限制） Content（含量） Identification（识别）,39,Quality Data Required at Method Validation（方法确认所需的质量数据）,40,Quality Data Required in Routine Analysis日常分析所需质量数据,1 may be required when working in unknown matrix,41,Review of
27、Quality Assurance质量保证概述,We will cover（内容） Definition and Description of QA 质量保证的定义和描述 Tools in a QA System（QA体系的工具） Proficiency Testing（能力验证） Certified Reference Materials（有证标准物质） Validated Methods（经过确认的方法）,42,Quality Assurance（质量保证）,Quality Assurance质量保证 Planned actions designed to provide confiden
28、ce that quality requirements will be fulfilled 为质量要求得到满足提供信心而设计的有计划的活动 Quality Assurance System质量保证体系 A set of processes applied within a defined structure to generate desired outcomes 在规定的结构内产生预期结果所采用的一组过程,The good management of good science良好的科学良好的管理,43,Quality System Components质量体系组成,Management管理
29、 Plan策划 Performance实施 Do做 Assessment评价 Review评审,44,6 Pillars of QA/QCQA/QC的六个基础,Management管理 Facilities设施 Personnel人员 CRMs有证标准物质 PTs能力验证 Validated Methods经确认的方法,45,Proficiency Testing能力验证,Provides external assessment of performance 为表现提供外部评审 Provides a basis for corrective and preventive action 为纠正和
30、预防措施提供基础,46,Information Gained搜集信息,Comparison of performance against an external standard 将表现与外部标准比较 Comparison against past performance 与过去的表现比较 Comparison against peer (competitor) labs与同行（竞争者比较） Identification of unsatisfactory performers识别出不满意的操作者,47,Types of PT能力验证类型,Specific analyte/matrix com
31、bination特定被分析物/基体结合 Mercury in fish（鱼中的汞） Benzene in air（空气中的苯） Field of analysis（场地分析） Metals by ICP（ICP测金属） Pesticides by GC/MS（ GC/MS测杀虫剂）,48,Samples样品,Sub-sampled from bulk supply 从批中抽取样本 Discrete sample split and distributed 谨慎分割样品并分发,49,Interlaboratory Comparisons实验室间比对,Collaborative studies 协
32、同研究 Certification studies 认证研究 Round-robin studies 循环研究,50,Selection of PT programsPT方案的选择,Availability of relevant programs 是否有相关方案 Organization of relevant programs 相关方案的组织 Any program is usually better than nothing 只有要方案，总比没有好,51,Assessment of PT ProgramsPT方案的评价,Before committing long-term consid
33、er 在实施前应考虑 Statistics used to interpret data 解释结果的统计技术 How “true” value is determined真值如何确定 Theoretical理论值 Referee lab参考实验室 Consensus公议值,52,Scoring of Participants参加者的评价,Z score（Z值） Deviation from “true” value divided by sigma 测得值与真值之差除以 Sigma may be an accepted deviation or statistically determined
34、 in each round 可能是可接受的偏差或每轮测试的统计结果,53,ScoringZ-值,Generally a Z-score （Z-值） 2, 3 is considered unsatisfactory（离群） scores must be within 2 sigma Z-值必须在2- 之间,54,Laboratory Interpretation实验室的解释,View results in context（从整体上看结果） Do they agree with in-house QC data 是否与室内质量控制数据一致 Trend analysis (consider ho
35、w sigma is determined) 趋势分析（考虑是如何确定的） Performance of other participants 其它参加者的表现,55,Laboratory Interpretation实验室的解释,Compare methodology and re-calculate if possible 比较方法，如果可能重新计算 Compare uncertainty against in-house estimates 将不确定度与室内评估比较,56,Corrective Actions纠正措施,Determine when to take action 确定何时采
36、取措施 After 1 unsatisfactory score? 一个不满意结果 After many (3) in one round? 一轮中多个（如三个）不满意结果 After 2 consecutive unsatisfactory scores? 连续二个不满意结果,57,Corrective Actions纠正措施,Always investigate总要调查 Include包括 QC data质量控制数据 Calculations, dilutions, transcriptions 计算、稀释，抄写 Training status of analyst分析人员培训状况 Per
37、formance records of equipment 设备运行记录 Records of calibrants, reagents etc. 校准品、试剂的记录等,58,Corrective Actions纠正措施,Formulate action公布措施 Execute action实施措施 Check if action was successful 检查措施是否有效,59,General Considerations几点考虑,More emphasis is being placed on PT programs by assessors, regulators and clien
38、ts 评审员、管理部门和客户更多强调PT Use all the information available 利用所有可得到的信息 Handle PTs like routine samples (as much as possible) 象常规样品一样处理PT样品（尽可能）,60,Third Party Assessment第三方评审,Accreditation Bodies认可机构 Assess performance against laboratory scope rather than PT program criteria 在一定范围内评审实验室的表现，而不是PT方案准则 Cons
39、ider “fit for purpose” 考虑适宜性,61,Third Party Assessment第三方评审,Regulators政府管理部门 May be involved in development of PT programs 可能参与PT方案的开发 May require assistance from program coordinator to assess results 可能需要方案协调人来帮助评审结果,62,Third Party Assessment第三方评审,Customers客户 When using PT results as a performance
40、standard ensure target is realistic 当使用PT结果作为能力标准时，确保目标是现实的 Must investigate program; not just results 必须调查方案，而不仅是结果,63,Reference Materials参考物质,5 types（5种类型） Pure substance（纯物质） Standard solutions and gas mixtures（标准溶液和气体混合物） Matrix RMs（基体参考物质） Physical properties（物理特性） Reference objects (artifacts)
41、（参考目标）（参考品）,64,Reference Materials参考物质,May be used for 可用于 Internal QC内部质量控制 Method validation方法确认 Calibration校准 Estimating uncertainty评估不确定度 Training培训,65,Classification分类,Certified Reference Materials有证参考物质 Directly traceable直接溯源 Each property has an uncertainty associated with it每一特性有相应不确定度 Refer
42、ence material参考物质 Properties are homogenous and thoroughly characterized to be used for calibration or method assessment 特性是均匀的，并完整定值，以用于校准或方法评价,66,Availability可获得性,Generally poor therefore expensive(不容易获得,费用高) Government agencies政府机构 NIST, NRC, EU-BCR Trade associations贸易协会 AOAC, AOCA (Oil Chemists
43、)石油化学 Commercial suppliers商业供货商,67,RM in Validation方法确认中参考物质的使用,Estimation of bias between method and true value评估方法和真值间的偏差 Estimation of bias between two methods评估方法间的偏差 Requires less rigorously certified RM 要求的有证标准物质相对不是太严格 Limited by uncertainty of certified values 受参考值不确定度的限制,68,RM in Validation
44、参考物质在方法确认中的应用,Matrix RM must be within scope of the method参考物质的基体必须适用于该方法 Multiple RMs may be required to cover range of method含多种成分的参考物质应适用于该方法,69,RM in Calibration参考物质在方法校正中的应用,Usually use pure substance常用纯物质 Uncertainty associated with purity will contribute to uncertainty of the method与纯度相关的不确定度
45、是方法不确定度的分量 Some methodology requires RM for calibration of entire process (XRF)有些方法学需要用参考物质对整个过程校准 (XRF),70,RM in Training参考物质在方法培训中的应用,Qualifying new analysts培养合格的新分析人员 Checking methods used infrequently检查不常用的分析方法 Troubleshooting methods解决分析方法中的疑难问题,71,RM as QC参考物质用于质量控制,Must be characterized with
46、respect to stability and homogeneity必须稳定、均匀,认定特征值,72,Suitability适用性,Matrix match (interferences)基体匹配程度 (干扰) Concentration range浓度范围 Uncertainty of certified values参考值的不确定度 Source (conformance with standards)来源 (与标准的符合性) Ongoing availability目前是否可得到 But How to assess但是 如何进行评估呢?,73,In-house production内部研究,What would the requirements be需要具备哪些必要条件?,