Supplement to Chapter on Risk Assessment

Presentations

  1. Lecture on risk assessment Slides►  Listen►
  2. Scoring for Multi-Morbidity Indices  Video► Slides►
  3. Tree diagrams & Bayes’ theorem   Slides►  Video► You Tube►
  4. Calculation of sensitivity, specificity, and Area under the Receiver Operating Curves Excel►
  5. Predicting outcomes Slides►  Video►
  6. Product of values of data in one column Slides► Video►
  7. Accuracy of Predictive Models (use instructor's last name as password) Read►
  8. Constructing Receiver operating Curves Slides► Video►

Assignments

Question 1: Construct a simple Multi-Morbidity Index.  Assess the average severity of CHF, MI, Diabetes, Hypertension, Alcohol Use, and ACL surgery (assume that sicker patients have longer stays). To calculate the average severity associated with a disease, compare all cases with the disease to all control patients without the disease. Make sure that in each comparison, patients with and without disease have the same set of comorbidities. For example, to find the average length of stay for patients with MI, select all MI patients with the following comorbidities: CHF,DM,AA.  Then compare these cases to controls who do not have MI but have the same comorbidities. To help you understand this assignment, consider the following table.  In this table, we see different strata of mutually exclusive and exhaustive comorbidities.  Then, within each strata we can observe the impact of MI.  The impact of MI is the average impact of MI within each strata.  Your objective is to create this table before you calculate the impact of MI.  To do so, first estimate n1 through n5 through a file where the data is restricted to cases with MI (WHERE MI=1).  Then, estimate the values n6 through n10 through a file where the data are restricted to non-MI patients (WHERE MI=0).  Merge these data, making sure that you match on the strata.  Then you can calculate the impact of MI.  Data►  Video► Slides►  SQL Code► Answer►

Strata Cases of MI Controls with No MI Difference
AA n1 n6 n1-n6
AA, CHF n2 n7 n2-n7
AA,DM n3 n8 n3-n8
DM, CHF n4 n9 n4-n9
AA,DM,CHF n5 n10 n5-n10

Question 2:  In question 1, calculate the likelihood ratio associated with each diagnosis in predicting above or below average length of stay.  First calculate the average length of stay for each diagnosis.  Assign individuals who have above average length of stay 1 and those below average length of stay 0.  Next calculate the likelihood ratio.  To calculate the likelihood ratio, select all individuals who have above average length of stay.  Examine the prevalence of the diagnosis among them.  Select all individuals who have below average length of stay and select the prevalence of the diagnosis among them.  The ratio of these two calculated numbers constitutes the likelihood ratio. We have calculated this number to be 1.67.  See if you get the same answer. Show intermediary calculations. Data► Access►

Question 3: Calculate the average age and the likelihood ratio associated with diagnoses. You would need to use SQL to do this assignment.  You can use any SQL software, including Access. Since the data is massive (17 million rows), keep in mind that Access requires you to analyze the data in partitions. Microsoft SQL server can analyze the entire data in one run. Submit your SQL code and the 10 diagnosis with highest and lowest likelihood ratios. For password to access data contact your instructor.  By opening this file you agree not to share the file with anyone else. Massive Data► Access► SQL Code► Marla's Guide► Answer►

Question 1: Use the attached data to create a receiver operating curve.  The file contains two values, predicted probabilities and actual true classification.   (a) Generate cutoff values as the average of two consecutive predicted values.  (b) Classify the model predictions.  (c) Calculate the sensitivity and specificity of model predictions at each cutoff level and list in order of the cutoff values.  (d) Draw the receiver operating curve.  (e) Calculate the area under the receiver operating curve.  Data► Curve► SQL►

More

  • Introduction to Standard Query Language More►
  • A comparative study of various severity measures. PubMed►
  • Use of severity of illness to classify and monitor medication errors More►
  • Use of pharmacy data to measure severity of illness More  
  • Risk of risk assessment More►
  • Nursing severity indices More►
  • Community wide measures of quality of care More►
  • Measures of outpatient severity of illness More►
  • Severity of episodes of illness More►
  • Disease Staging More►
  • Patient Management Categories More►
  • The Acute Physiological and Chronic Health Evaluation (APACHE) index More►
  • Medisgroup More►
  • Computerized Severity Index More►
  • Indices that rely on select categories of diagnoses  Charlson Index► Elixhauser List►
  • Comorbidity-Polypharmacy Score More►