Blok AWV2 2020/2021 UL

Deze bundel bevat alle aantekeningen van de colleges uit het blok AWV uit het 2e jaar van de bachelor Geneeskunde aan de Universiteit Leiden. Ook aantekeningen uit de werkgroepen zijn in de samenvattingen verwerkt.

Bundle items:

Blok AWV HC1: Research questions

The ultimate goal of medical research is to improve medical practice. There are different aspects of medical practice which form 4 classes of medical research:Etiology (prevention/risk factors of a medical disease)Is a high caloric diet a risk factor for cardiovascular disease?DiagnosisWhat is the probability of having a hip fracture if the affected leg is shorter and in exorotation?TreatmentDoes chloroquine treatment reduce the risk of mortality among COVID-19 patients admitted to the ICU?PrognosisWhat is the probability of dying within 5 years after breast cancer diagnosis? Research questions arise in practice. It is important to particularly articulate a research question → they are the starting point when designing a study or when reading a paper. A research question should be answerable and have standard elements/components. Research questions can consist of several types of standard components:PICO: most often usedPatient (population)InterventionComparatorOutcomeDDODomainDeterminantOutcome Examples of research questions are:To what extent does alemtuzumab compared to interferon-β improve the time until relapse occurs in patients with relapsing remitting MS?Classification: treatmentPatients: patients with relapsing remitting MSIntervention: alemtuzumabComparator: interferon-βOutcome: time until relapse occursWhat is the predictive value of a second ultrasound, in case a first ultrasound was inconclusive, in patients suspected of having an acute appendicitis?Classification: diagnosisPatients: patients suspected of having an acute appendicitis, in whom a first ultrasound was inconclusiveIntervention: second ultrasoundComparator: /Outcome: appendicitisIs smoking a risk factor for lung cancer?Classification: etiologyPatients: humansIntervention: smoking...

Read more about Blok AWV HC1: Research questions

Blok AWV HC2: RCT

ReDuCe is a new diuretic drug for patients with hypertension. Its effect has to be studied → a research question has to be made:What is the effect of a new diuretic drug (ReDuCe) versus no diuretic treatment on blood pressure in patients who have an increased blood pressure?Classification: treatment researchPatients: subjects with an increased blood pressureIntervention: 40 mg ReDuCeComparator: no diuretic treatmentOutcome: blood pressure reduction After 6 weeks of treatment, a patient prescribed with ReDuCe has a normalized blood pressure → the effect. However, this effect can have several causes:Start of the pharmacological therapy (ReDuCe)Stopping of smokingStarting to go to the gymMore adherence to β-blocker treatmentFear of the doctor the first time the blood pressure was measuredThis is called “white coat hypertension”Use of a more precise measurement instrument the second timeThe blood pressure fluctuation over timeRegression to the mean has to be appliedSometimes the blood pressure is relatively high, sometimes relatively lowEt cetera The outcome is a combination of 4 phenomena:Treatment (T)Natural course (NC)Natural reasons for why the blood pressure goes up or downExtraneous factors (EF)E.g. going to the gymError processes (V)Variation of the measurement deviceTherefore, outcome = T + NC + EF + V. The interest mainly lies in the effect of the treatment (T):Outcome with treatment : T + NC + EF + VOutcome without treatment : NC + EF + V To identify the effect of treatment, 2 (or more) groups need to be compared. These groups should be comparable with respect to NC, EF and V, and differ only with respect to treatment. In this case, an observed difference in the outcome between the groups can be attributed to the only aspects that the 2 groups differ on → the treatment. Comparability is necessary....

Read more about Blok AWV HC2: RCT

Blok AWV HC3: Sample size calculation

In medical papers, there often is a statistical analysis paragraph with a motivation of the number of people in a sample of a study.The aim of an RCT is to compare 2 treatments → patients are recruited to the study and randomized to treatment A or B. It also needs to be determined how many patients are included in the RCT → the sample size:Too fewImprecise results → no power to determine the effect of treatmentToo manyTakes a lot of time, effort and moneyIt is unethical to include many patients in a study Factors for deciding sample size are:PracticalNumber of eligible patients treated at the centerNumber of patients willing to participateTimeMoneyStatisticalHow big of an effect can be detected with a given number of patients? Hypothesis testing yields P-values and statements of statistic significance. Hypothesis testing is done as follows:Decide on a null hypothesis (H0) about the populationH0: there is no difference between the 2 groupsTake a representative sample of the populationCalculate the observed difference in the sampleCalculate the p-valueP-value: the probability to observe at least this difference if H0is trueThis is done by a statistical testIf the p-value is smaller than the prespecified value α, H0is rejectedThe value αis called the significance level However, mistakes can be made in hypothesis testing. H0is rejected in case the observations are unlikely to occur if H0is true, and not if they are impossible:Correct decisionsH0is not rejected + H0 is trueH0is rejected + H0is not trueIncorrect decisionsH0is rejected + H0is true → a type 1 errorα = the probability of a type 1 errorH0is not rejected + H0is not true → a type 2 errorβ = the probability of a type 2 errorPower = 1 - β The power is the probability of finding a...

Read more about Blok AWV HC3: Sample size calculation

Blok AWV HC4: Cohort studies

Medical research almost always requires collection of data. Study design described how is the study set up and how the data are collected. There are 2 types of study design:Cohort studyCase-control study A cohort is a group of people with a common characteristic (gender, diabetes, age, etc.). A cohort study follows such a group over time and records certain outcomes in a follow-up study or longitudinal study. Usually, 2 or more groups are compared. For instance, with patients who are hospitalized for Covid-19, there is a follow-up, record of the mortality and comparison of chloroquine treatment versus no chloroquine treatment. Cohort studies are often simplified in tables. An example is the association between smoking and lung cancer:Cases: individuals who develop of an outcome of interestN = size of the groupT = person timeIf a person contributes 50 weeks to the study, and another person 2 weeks, the person time is 52 weeks → 1 person year CasesNExposedAN1UnexposedBN0 CasesNSmokers1600100 000Unexposed400200 000 CasesTExposedAT1UnexposedBT0 CasesNExposed1600300 000Unexposed400600 000With a cohort, the risk/rate can be estimated:Cumulative incidence = A/N1→ 1600/100 000 = 1,6%The cumulative is also known as the risk → the risk of smokers getting lung cancer is 1,6%Time is not taken into accountIncidence rate = (A/T1) = 1600/300 000 years = 5,3 per 1000 person yearsTime is taken into accountIt is expected that 5,3 per 1000 individuals develops lung cancer in 1 yearRisk ratio (RR) = (A/N1)/(B/N0) → (1600/100 000)/(400/200 000) = 1,6%/0,2% = 8Can be either the cumulative incidence ratio or the incidence rate ratioCumulative incidence risk ratio = (A/N1)/(B/N0)Incidence rate ratio = (A...

Read more about Blok AWV HC4: Cohort studies

Blok AWV HC5: Case control studies

In a case control study, instead of a census (information about everybody), sampling is done. The cases are compared to a sample of the entire population. This happens as follows:Case detectionSample of the controlsExposure status assessment among cases and among controlsE.g. comparison between the smoking status among the population and the smoking status among the populationExposure-outcome relation estimation A case-control study is very efficient compared to a cohort study:Cohort: requires information about the entire study populationCase-control: requires information about part of the controlsAdvantage: practical efficiency, particularly when:Disease is rareThere is a long/unknown latency periodAssessment exposure is expensiveDisadvantage: it is not possible to estimate absolute disease frequencies (risk/rate)Case-control studies do not allow for:RandomizationAbsolute measures of disease frequency The odds ratio is the ratio of odds of exposure among cases and odds of exposure among controls. It is the only measure that can be estimated with a case control study.Case-controlCasesControlsExposedacUnexposedbdThe following calculations can be done:Odds ratio (OR) = (a/b)/(c/d) = ad/bc The interpretation of the odds ratio depends on how the control study is done and what the controls represent. How this can be interpreted depends on the moment of sampling controls. Moments of sampling controls are:At the end of the follow-up → sampling from non-casesAt the start of the follow-up → case-cohortDuring the follow-up (each time a case occurs) → incidence density samplingEach time a case is identified, a control is sampled from the cohortControls can represent various things, such as:Non-casesTotal number of subjectsPerson...

Read more about Blok AWV HC5: Case control studies

Blok AWV HC6+7: Bias

Medical research often consists of the quantification of a phenomenon, for example quantifying the effect of cholesterol lowering treatment on the risk of cardiovascular events among those with elevated serum cholesterol levels. Ideally, results apply not only to those who participate in the study (the study population/sample), but also to others with elevated cholesterol levels (the domain).The domain consists of a large group for whom the results of the study should apply. Of this domain, a sample is selected based on:In-/exclusion criteriaRegionTime periodEt ceteraWhile the study is done within this sample, the results of the study should apply to the domain. If there is a systematic error, results of the study do not apply to the domain. Systematic errors in design, conduct or analysis may lead to erroneous results → bias. An example of a systematic error is a broken measuring tape when measuring the length of something. Each time the measuring tape is used, results are incorrect → a systematic error is not solved by a large sample size. Random errors can be decreased by increasing the sample size. As the study size increases, random errors start playing a increasingly smaller role. There are 3 sources of bias:Measurement errors → information biasMissing data → selection biasIncomparability of study groups → confoundingThere can be bias in different parts of studies, depending on the study type:Randomized trialOutcomeCohort studyExposureOutcomeConfoundersCase-control studyExposureOutcomeConfounders Examples of measurement errors are:Questionnaires for measuring body weight instead of scalesDrug use measured by “prescription information” instead of controlling if the drugs actually are takenMeasuring smoking status with “yes/no” instead of pack years...

Read more about Blok AWV HC6+7: Bias

Blok AWV HC8+9: Survival analysis

Survival is used to determine when a certain event happens, such as:When a tumor develops after exposure to a carcinogenWhen a patient dies after a cancer diagnosisWhen a patient becomes pregnant after the start of fertility treatmentWhen a graft is rejected after transplantationPossible questions that can be answered with survival analysis are:What is the distribution of the survival times?Is there a difference in expected life-time between different treatments?What is the fastest way to become pregnant?Which factors predict 5-year survival probabilities? For patients with end stage renal disease, it needs to be determined which treatment gives better survival:Hemodialysis (HD)Peritoneal dialysis (PD)Survival probabilities and survival times after start of dialysis can possibly be compared with help of T-tests of Chi square tests. However, there are several problems:Not all patients diedNot all patients have the same follow-up timeSome started dialysis a while ago, some started dialysis recentlyPatients get lost (e.g. migration) Because the time of death of all patients isn’t known, some patients have censored survival times. In an ideal world, patients that are still alive in each group are compared after a certain time period. However, some patients may be lost to follow-up. This can be solved by censoring survival times → the time of event (e.g. death) is not observed, it is only known when the patient was last seen alive. Reasons for censoring are:Administrative censoring: individuals do not have the event (death) before the end of the studyLost to follow-up: the patient moves or does not show up for appointmentsThe patient dies of another disease In the peritoneal dialysis group, 207 patients died and 446 are still alive and were last seen between 0,8 and 5 years after the start of...

Read more about Blok AWV HC8+9: Survival analysis

Blok AWV HC10+11: Regression analysis

Statistics consists of making statements about a population based on data observed from a sample. This is often done using means and standard deviations (σ). The bigger the standard deviation, the bigger the spread in the population.For example, the lung function (FEV1 in L) of 40 children is measured:Mean FEV1 = 3,16 Lσ = 0,41 LThis means that roughly 95% of the population has a FEV1 between 3,16 – 0,82 and 3,19 + 0,82 L → approximately 95% of observations are less than 2σ from the mean:95% CI = (2,34 L, 3,84 L)However, lung function depends on many factors such as age and gender. These factors also need to be taken into account. Simple linear regression is regression for continuous outcomes. Linear regression tries to predict or explain a variable → the outcome or the dependent variable (x). This variable is explained by another variable → the explanatory variable (y). A regression line is based on a scatter plot and calculates the mean value of “y” for a value of “x”:y = the dependent variable, outcome and response variablex = the independent variable, covariate, risk factor, predictor and explanatory variableMean y = β0+ β1xβ0= the intercept (“constante”)The predicted value of “y” if “x” is equal to 0Not always clinically meaningfulβ1= the slope (“richtingscoëfficiënt”)The expected change in the outcome by increasing the exposure of 1 unit if β1 is positiveOr decrease, in case β1is negativeFor instance, a regression line can describe the mean FEV1 as function of age:Mean FEV1 = 2,281 + 0,119 x ageThis means that for 2 children with an age difference of 1 year, the expected mean difference in the FEV1 is 0,119 L. Observations of (x1, y1), (x2, y2), …, (xn, yn) show that each pair represents the values of 1 person. Sometimes, the error can also be taken into account:y = β0+...

Read more about Blok AWV HC10+11: Regression analysis

Blok AWV HC12: Diagnostische begrippen

Een kans is een getal tussen 0 en 100% dat weergeeft hoe waarschijnlijk iets is. Deze kans stijgt als een test positief is, en daalt als een test negatief is. Belangrijke termen zijn:Voorafkans/a-priori kans/prevalentie: voor een testAchterafkans/a-posteriori kans/voorspellende waarde: na een testSensitiviteit: kans op een positieve test als de ziekte aanwezig isSpecificiteit: kans op een negatieve test als de ziekte afwezig isPositief voorspellende waarde: kans dat de ziekte aanwezig is als de test positief isNegatief voorspellende waarde: kans dat de ziekte afwezig is als de test negatief is Op een polikliniek chirurgie komt een jonge patiënt, die sinds een halve dag buikpijn heeft. De pijn is heftig, continue en gelokaliseerd op het punt van McBurney. De dokter denkt aan appendicitis acuta:20% van alle acute buik patiënten heeft een appendicitis acuta → de voorafkans is 20%Van alle acute buik patiënten met appendicitis acuta zegt 90% pijn te hebben op McBurney → de sensitiviteit is 90%Van alle acute buik patiënten zonder appendicitis acuta zegt 15% pijn te hebben op McBurney → de specificiteit is 85%Positieve testuitslagen kunnen terecht positief (TP) of fout positief (FP) zijn:TP = voorafkans op ziekte x sensitiviteit → 20% x 90% = 18%FP = voorafkans op geen ziekte x (1 – specificiteit) → 80% x 15% = 12% Deze gegevens kunnen gezet worden in een 2x2 tabel: AppendicitisGeen appendicitisTotaalWel pijn MB18%12%30%Geen pijn MB2% (0,1 x 0,2)68% (0,85 x 0,8)70%Totaal20%80%100%De achterafkans na een positieve test is 18% van 30% → 60% kans. De kans dat de patiënt appendicitis heeft is 60%. Dit heeft ook wel de positief voorspellende waarde:VW+ = TP/(TP + FP) = 18%(18% + 12%) = 60%De negatief voorspellende waarde is de kans dat ziekte afwezig is, gegeven de negatieve...

Read more about Blok AWV HC12: Diagnostische begrippen

Blok AWV HC13: Beslisbomen

Een beslisboom bestaat uit takken met beslisknopen en eindknopen. Een voorbeeld is het drie deuren probleem, waarbij achter 1 deur een auto staat. Vervolgens wordt van 1 deur bekend wordt dat deze fout is waardoor de kans 50% is dat achter 1 van de overgebleven deuren het goede zit. Vervolgens kan de persoon die voor 1 van de overgebleven 2 deuren staat besluiten om wel of niet te wisselen. Voor beide opties zijn 2 mogelijkheden:Niet wisselenEerst goede deur → kans van 1/3 → de auto wordt gewonnenEerst foute deur → kans van 2/3 → niks wordt gewonnenWel wisselenEerst goede deur → kans van 1/3 → niks wordt gewonnenEerst foute deur → kans van 2/3 → de auto wordt gewonnenDit zijn de kansen vóór het wisselen. Deze analyse laat zien dat in het geval dat er niet gewisseld wordt, de kans op het winnen van de auto 1/3 is, terwijl als er wel gewisseld wordt, de kans 2/3 is. Spijt ontstaat als de persoon eerst goed stond, en daarna wisselde waardoor de auto alsnog niet is gewonnen. Een aantal opmerkingen bij het 3-deuren probleem zijn:De goede beslissing geeft niet steeds de beste uitkomstProces- versus uitkomstkwaliteitEffectiviteit versus bijwerkingenHet beslissingscriterium bepaalt de optimale beslissingWat is het doel?Vooral de auto willen hebben of geen spijt willen hebben?Geanticipeerde spijt → achteraf geen spijt willen hebben Met beslisbomen kunnen problemen geanalyseerd worden:StructuurKeuzeknopen, kansknopen en eindknopenVan links naar rechtsGetallenGetallen bij de kansknopenDe kans op een hele “tak” is p1x p2x …Uitkomsten staan bij de eindknopenAnalyseren en optimaliserenBereken de verwachtingswaarde voor elke beslissingKies d ebeste verwachte waarde Over het algemeen is de verwachtingswaarde ongeveer “het midden” → in een continue of normale verdeling...

Read more about Blok AWV HC13: Beslisbomen

Blok AWV HC14: Test en behandeldrempel

De behandeldrempel beantwoord de klinische vraag: “Hoe zeker moet ik zijn van de ziekte, voordat ik overga tot behandeling, wetende dat elke behandeling zowel voor- als nadelen heeft?” Elke behandeling heeft voor- en nadelen. De consequenties van een behandeling bepalen wanneer een behandeling wordt ingezet → er wordt gekeken naar de balans tussen de voor- en nadelen:Baat: vermindering van de schade door de ziekteKan alleen optreden bij de aanwezigheid van die ziekteKosten: gezondheidsschade door de behandeling zelfBij mensen met en zonder de ziekteBij mensen met de ziekte, zitten de kosten al verwerkt in de baatMensen zonder de ziekte hebben alleen de kostenBaat het niet, dat schaadt het nietHet volgende geldt:Bij zekerheid over de ziekteBij mensen met de ziekte is er baatBij mensen zonder de ziekte zijn er kostenBij onzekerheid over de ziekteBij mensen met een hoge kans op de ziekte is er meer baat dan kostenBij mensen met een lage kans op de ziekte zijn er meer kosten dan baatBij de behandeldrempel is de baat gelijk aan de kosten. De baat en kosten zijn ook zichtbaar in de beslisboom, bijvoorbeeld over AAAA:OpererenDe patiënt heeft AAAA → overlevingskans van 40%De kans hierop is pAAAADe patiënt heeft iets anders → overlevingskans van 95%AfwachtenDe patiënt heeft AAAA → overlevingskans van 0%De kans hierop is pAAAADe patiënt heeft iets anders → overlevingskans van 100%In dit geval is de baat 40% - 0% = 40%, en zijn de kosten 100% - 95% = 5%. Elke beslissing heeft zijn eigen uitkomst → elke uitkomst heeft een “utility” (waarde of nut):UTP= nut van “true positive” → terecht behandeldUFP= nut van “false positive” → onterecht behandeldUFN= nut van “false negative” → onterecht onbehandeldUTN= nut van “true negative” → terecht onbehandeldHiermee kunnen de...