Re-print changes 2001.pdf

Constructing Written Test Questions
For the Basic and Clinical Sciences
Third Edition
National Board of Medical Examiners Chapter 6
Extended-Matching (R-Type) Items
Extended Matching items are multiple-choice items organized into sets that use one list of options for all items in the set.
A well-constructed Extended-Matching set includes four components: 1. a theme; 2. an option list; 3. a lead-in statement; and 4. at least two item stems, as illustrated below. Options:
Lead-in:
For each patient with fatigue, select the most likely diagnosis. 1. A 19-year-old woman has had fatigue, fever, and sore throat for the past week. She has a temperature of 38.3 C (101 F), cervical lymphadenopathy, and splenomegaly. Initial laboratory studies show a leuko-cyte count of 5000/mm3 (80% lymphocytes, with many lymphocytes exhibiting atypical features).
Serum aspartate aminotransferase (AST, GOT) activity is 200 U/L. Serum bilirubin concentration andserum alkaline phosphatase activity are within normal limits.
2. A 15-year-old girl has a two-week history of fatigue and back pain. She has widespread bruising, pallor, and tenderness over the vertebrae and both femurs. Complete blood count shows hemoglobin concentra- tion of 7.0 g/dL, leukocyte count of 2000/mm , and platelet count of 15,000/mm .
Chapter 6. Extended-Matching (R-Type) Items
Extended-matching items are written differently than traditional one-best-answer items. Most often, the theme, lead-in, andoptions are written first; the item stems are written last. For example, if you want to write some questions related to the diag-nosis of fatigue, you would begin by listing the diagnoses that might cause fatigue. You would then write a vignette for each(or many) of the options in the list. The example above includes vignettes for Epstein-Barr virus infection and for acuteleukemia. Additional items might be written for some of the remaining diagnoses; for common, treatable diagnoses, morethan one item might be prepared. The sample vignettes are moderate in length; shorter, more-focused vignettes could alsobe used. Alternatively, examinees could be challenged to identify key diagnostic information, intermingled with incidentalfindings, by using longer vignettes. Avoiding Flaws When You Write Extended-Matching Items for Your Own Examination
The four components (theme, options, lead-in, and stems) are all essential for the construction of a good quality extended-matching set. Sets without lead-ins (or with nonspecific lead-ins, such as “Match each item with the best option”) should NOTbe used, because they generally pose inconsistent or ambiguous tasks for examinees. The following set is flawed. The optionsare heterogeneous; there is no lead-in; the stems cannot be answered without reading the options. Rules for extended-match-ing items are completely analogous to those for one-best answer items.
Sample Extended-Matching Set - Flawed
Factors that people consider when evaluating air quality A principal effect of particulate matter in air After reading the stem in Item #1, examinees have only the vaguest idea what the question is about. In an attempt to deter-mine the “best” answer, the examinees have to decide whether “is motion sickness” is more or less true than “have no effectson people.” The task is not do-able. Under these circumstances, unless an option is absolutely 100% true or false, it cannotbe rank-ordered with the other options. The stem of item #1 by itself is not clear; the item cannot be answered without look-ing at the options.
As with one-best-answer items, the stems should be long; the options should be short. There MUST be a lead-in that estab-lishes the relationship between the items and the options. There should be NO verbs in the options. The “cover-the-options”rule is as relevant for extended-matching items as it is for one-best-answer items.
Chapter 6. Extended-Matching (R-Type) Items
Sample Lead-ins and Topics for Option Lists
Patient vignettes provide an excellent structure for stems, not only in the clinical sciences, but also to assess knowledge inthe basic sciences. Lead-ins generally begin with a phrase such as “For each of the following patients.” Often sets areorganized around chief complaints or some other factor that allows a more specific introductory phrase such as “For eachof the following patients with fatigue,” or “For each of the following patients with an enzyme deficiency.” The second partof the lead-in describes the task and the option set: “select the most likely diagnosis”; “select the protein that is most like-ly to be defective.” The following are some additional sample lead-ins and some suggested topics for option lists.
For each of the following patients, select the [eg, nerve] that is most likely to be [abnormal/defective/deficient/non-functioning].
Options sets could include list of nerves; list of muscles; list of enzymes; list of hormones; list of proteins; list oftypes of cells; list of neurotransmitters; list of pathologic processes. For each of the following patients, select the [finding] that would be expected.
Options sets could include list of laboratory results; list of additional physical signs; autopsy results; results of micro-scopic examination of fluids, muscle or joint tissue; DNA analysis results; hormone levels. For each of the following patients, select the most likely [cause].
Options sets could include list of underlying mechanisms of the disease; medications that might cause sideeffects; list of drugs or drug classes; toxic agents; hemodynamic mechanisms. For each of the following patients, select the [eg, drug] that should be administered.
Options sets could include list of drugs, vitamins, amino acids, enzymes, hormones. For each of the following patients with [chief complaint], select the most likely diagnosis.
Options sets could include list of diagnoses, most often organized around a chief complaint such as diseases thatcause chest pain or diseases that cause fever. For each of the following patients, select the most appropriate next step in patient care.
Options sets could include list of pharmacologic therapies, list of laboratory studies, disposition alternatives, orthe options could include a mixed set of treatments and additional studies to assess whether the student knowswhen sufficient data have been gathered. More on Options for R-Sets
Generally, anything that can be listed can form the basis for options in an R-set. Below are some topics that have been usedas the basis for option lists.
The list of options should be single words or very short phrases. They must be homogeneous (all diagnoses, all manage-ment options, all anatomical sites, all vitamins, etc). They can be labeled areas in a graph or in pictorial material. Options,especially those involving laboratory values, are often expressed in tabular form (see physiology example). Include all rel-evant options that are appropriate for the examinees; subtle distinctions and uncommon diagnoses may be inappropriate.
For some topics, as few as three options might be appropriate; for others, a list of 26 (one for each letter in the alphabet)might be required. Chapter 6. Extended-Matching (R-Type) Items
Writing the Item Stems
Patient vignettes provide an excellent structure for stems both in the basic and the clinical sciences. In the clinical sciences,the vignette commonly provides the patient’s age, gender, chief complaint, and site of care, followed by personal history, fam-ily history (if relevant), then physical examination information, then laboratory data (if provided). Depending upon the pur-pose of the set, vignettes can be brief, prototypic presentations or fuller descriptions that challenge examinees to identify keyinformation. Generally, these items would include at least the patient’s age, gender, chief complaint, and related history. Itemsto assess knowledge of the basic sciences, particularly for courses taught in the first year of medical school, might include lessdetailed prototypical descriptions. Each patient description should be similar in structure to the others in the set. For example, if race, ethnicity, or occupation isincluded in one item, it should be included in all items; if laboratory data are included for one item, include them in all items.
It is advisable not to mix adults and pediatric cases in the same set — too often the age alone provides too much cueing andeliminates large numbers of options from consideration. One advantage of the use of patient vignettes is that it helps to assure that the content assesses application of knowledge. Theseitems should not resemble crossword puzzles, where both the options and the stems are single words or short phrases. Avoidreconstructing those items you were faced with in junior high school where you had to draw a line from something in columnA to the matching option in column B. It is particularly important that the items be straightforward. There is no reason to make them tricky; the extended option listmakes them difficult enough to allow you to distinguish the knowledgeable student from the other students without resortingto trickery. As with well-constructed A-type questions, the “cover-the-options” rule is paramount. Knowledgeable studentsshould be able to generate an answer to the question and then find that answer in the alphabetical list of options. An item should be prepared for most of the options; for common or important options, more than one item can be written. Inconstructing an examination assessing general competence, to avoid overemphasizing a topic, all of the options, but only twoor three of the items, would be used; the remaining items are retained for subsequent exams. On the other hand, if you wantto assess knowledge in greater depth on a smaller number of topics, 10 to 20 items can be included for each set, with a sub-score calculated for each topic.
In reviewing the items, check to make sure that there is only a single “best” answer for each question. Also make sure thatthere are at least four reasonable distractors for each item. As a final check, it is recommended that you ask a colleague toreview the items (without the correct answer indicated). If the colleague has difficulty determining the correct answer, mod-ify the option list or the item to eliminate the ambiguity.
Sample Good and Bad Item Stems Using the Same Option List
The following is a good microbiology set. The options are a homogeneous list of pathogens; including both viruses and bac-teria makes sense. There is a lead-in that presents a clear task for the examinee. There are two item stems that require stu-dents to apply their basic science knowledge of microbiology to arrive at the most likely cause of each patient’s illness. For each patient with fever, select the pathogen most likely to have caused his/her illness. 1. A 7-year-old girl has a high fever and a sore throat. There is pharyngeal redness, a swollen right tonsil with creamy exudate, and painful right submandibular lymphadenopathy. Throat culture on blood agar yields numerous small $-hemolytic colonies that are inhibited by bacitracin.
2. For the past week, an 18-year-old man has had fever, sore throat, and malaise with bilaterally enlarged tonsils, tonsillar exudate, diffuse cervical lymphadenopathy, and splenomegaly. There is lymphocytosis with atypical lymphocytes. Thepatient tests positive for heterophil antibodies.
The following stem, developed for the same set, assesses recall of isolated facts rather than application of knowledge. Itlooks more like a crossword puzzle question than a question for a medical school examination. 3. An encapsulated gram-positive organism that usually grows in pairs or short chains.
Chapter 6. Extended-Matching (R-Type) Items
The following set includes two item stems. The first requires that the examinee synthesize information in order to deter-mine a diagnosis; the second requires only recall of an isolated fact.
For each patient with clinical features caused by metabolic abnormalities, select the vitamin or mineral that is most likelyto be involved. 1. A 70-year-old widower has ecchymoses, perifollicular petechiae, and swelling of the gingiva. His diet consists most- 2. Involved in clotting factor synthesis.
The following set includes two item stems. The first item stem requires that the examinee synthesize information to deter-mine the diagnosis of trisomy 21; the second item provides this information. When you develop item stems, you need todecide the level of synthesis you will provide.
Complete transposition of the great arteries For each patient, select the most likely congenital heart defect. 1. A 3090-g infant is born at term. Physical examination shows the child to be hypotonic with a weak suck. There is no cyanosis. Prominent epicanthal folds, a large tongue, and small incurved 5th digits on the hands are present. Thereis a loud holosystolic murmur over the entire precordium and a palpable thrill at the upper left sternal border in the2nd through 4th intercostal spaces.
2. An infant with trisomy 21 has clefts of the right and left atrioventricular valves; there is no cyanosis.
Chapter 6. Extended-Matching (R-Type) Items
This set poses a task that is clinically backward. The examinee is given a vaccine and asked to select the best patient for thisvaccine. A more appropriate task would be for the examinee to be given a patient and asked for the most appropriate nextstep in patient care (ie, describing a patient in each item stem; using “For each patient, select the most appropriate next stepin patient care”; and using things such as vaccines as options). A second problem with the set is that insufficient informa-tion is provided about each patient. For example, the examinee would want the immunization history for a patient beforedeciding which immunizations to provide.
Occupation
Pregnant
Childhood
For each vaccination, select the patient profile that represents its most appropriate use. Overview of the Steps for Writing Extended-Matching Items
1. Identify the theme for the set. The theme can be a chief complaint (eg, chest pain, fatigue), a disposition situation
(eg, admission/discharge from the emergency department), a drug class (eg, antihypertensive agents, antibiotics). 2. Write the lead-in for the set (eg, For each patient described below, select the most likely diagnosis). The lead-in
indicates the relationship between the stems and options, clarifying the question posed for examinees. It is an essen-tial component of an Extended-Matching set.
3. Prepare the list of options. The list of options should be single words or very short phrases. List the options in
alphabetical order unless there is a logical order.
4. Write the items. The items within a set should be similar in structure. Most often, patient vignettes are appropriate.
5. Review the items. Check to make sure that there is only a single “best” answer for each question. Also make sure
that there are at least four reasonable distractors for each item. As a final check, it is recommended that you ask acolleague to review the items (without the correct answer indicated). If the colleague has difficulty determining thecorrect answer, modify the option list or the item to eliminate the ambiguity.
Additional information on writing Extended-Matching items can be found in: Case SM, Swanson DB. Extended-matching items: a practical alternative to free-response questions. Teaching and Learning in Medicine.
1993;5(2):107-115.
Case SM, Swanson DB, Woolliscroft JO. Assessment of diagnostic pattern recognition skills in medicine clerkships using a written test. In: Hard-en R, Hart I, Mulholland H, eds. Approaches to Assessment of Clinical Competence. Norwich, England: Page Brothers; 1992:452-458.
Chapter 6. Extended-Matching (R-Type) Items
Sample Extended-Matching Sets
Sample Anatomy Set
For each patient with neurologic abnormalities, select the artery that is most likely to be involved. 1. A 72-year-old right-handed man has weakness and hyperreflexia of the right lower limb, an extensor plantar response on the right, normal strength of the right arm, and normal facial movements. 2. A 68-year-old right-handed man has right spastic hemiparesis, an extensor plantar response on the right, and paral- ysis of the lower two-thirds of his face on the right. His speech is fluent, and he has normal comprehension of ver-bal and written commands.
Sample Pharmacology Set
For each patient, select the drug most likely to have caused the adverse effect. 1. A 56-year-old man with recurrent ventricular arrhythmias began taking an antiarrhythmic drug 5 months ago. He now has progressive dyspnea, cough, and low-grade fever. Erythrocyte sedimentation rate is increased. X-ray filmof the chest shows a diffuse interstitial pneumonia. Pulmonary function tests show that diffusing capacity for car-bon monoxide is decreased.
2. A 62-year-old man with chronic obstructive pulmonary disease begins therapy with an antihypertensive drug. Two weeks later, he has marked worsening of dyspnea and clearly audible wheezing.
Chapter 6. Extended-Matching (R-Type) Items
Sample Physiology Set — Clinical Features (Which additional findings are likely?)
HCO - mEq/L
For each patient described below, select the most likely arterial blood gas findings. 1. A 22-year-old man with a 3-week history of polyuria and polydipsia has had nausea, vomiting, and decreased respon- siveness for the past 12 hours. Urinalysis (dipstick) shows 4+ glucose and 4+ ketones.
2. A 25-year-old woman is brought to the emergency department 12 hours after a suicide attempt. She took approxi- Sample Diagnosis Set
For each patient with back pain, select the most likely diagnosis. 1. A 26-year-old man has insidious onset of low back pain and early morning stiffness. The pain alternates from side to side and occasionally radiates into the buttocks and back of the thighs, but not below the knees. The patient hasacute anterior uveitis, diffuse low back and sacroiliac tenderness, and restricted range of motion at the hips. His ery-throcyte sedimentation rate is 40 mm/h; latex fixation test is negative; and mild hypoproliferative anemia is present.
2. Twelve hours after being struck from the rear while driving her car, a 28-year-old woman has vague cervical and lumbar pain associated with headache and restricted cervical mobility. She is now very anxious. Rope-like bandsof muscle are present in the lumbar area and over the left buttock; the bands are painful Additional items would cover some of the remaining diagnoses. The sample vignettes are of average length; shorter, morefocused vignettes may also be used. Alternatively, examinees could be challenged to identify key diagnostic information byusing longer vignettes. Chapter 6. Extended-Matching (R-Type) Items
Sample Diagnosis Set
For each patient with abdominal pain, select the most likely diagnosis. 1. A 25-year-old woman has sudden onset of persistent right lower abdominal pain that is increasing in severity. She has nausea without vomiting. She had a normal bowel movement just before onset of pain. Examination showsexquisite deep tenderness to palpation in right lower abdomen with guarding but no rebound; bowel sounds are pre-sent. Pelvic examination shows a 7-cm, exquisitely tender right-sided mass. Hematocrit is 32%. Leukocyte countis 18,000/mm3. Serum amylase activity is within normal limits. Test of the stool for occult blood is negative.
2. An 84-year-old man in a nursing home has increasing poorly localized lower abdominal pain recurring every 3-4 hours over the past 3 days. He has no nausea or vomiting; the last bowel movement was not recorded. Examination showsa soft abdomen with a palpable, slightly tender, lower left abdominal mass. Hematocrit is 28%. Leukocyte count is10,000/mm3. Serum amylase activity is within normal limits. Test of the stool for occult blood is positive.
Sample Management Set: Disposition
For each of the following patients, select the most appropriate next step in patient care. Items might describe patients with appendicitis, ectopic pregnancy, endometriosis, Crohn’s disease, diverticulitis, pelvicabscess, sickle cell crisis, renal lithiasis, twisted ovarian cyst, or other problems that commonly present as emergencies.
Other disposition sets might focus on “telephone triage,” hospital transfer/discharge decisions, etc. Chapter 6. Extended-Matching (R-Type) Items
Sample Management Set: Diagnostic Testing
For each patient who comes to the physician for a health maintenance examination, select the most appropriate diagnostic study. 1. A 22-year-old man who weighs 89 kg (196 lb) and is 175 cm (69 in) tall has smoked one pack of cigarettes daily for 8 years; he does not exercise. His last examination was 5 years ago. His father had a myocardial infarction at theage of 48 years. Physical examination shows no abnormalities. 2. A 28-year-old woman who weighs 70 kg (154 lb) and is 173 cm (68 in) tall has smoked one pack of cigarettes daily for 12 years; she does not exercise. Her last examination was 5 years ago, though she had a Pap smear 9 months agothat showed normal results. Her father had a myocardial infarction at the age of 48 years. Her grandmother wasdiagnosed with colon cancer at the age of 62 years. Physical examination shows no abnormalities.
Sample Option List for Electrolyte Abnormalities
For each of the following patients, select the electrolyte abnormality most likely to be present. Sample Behavioral Sciences/Pediatrics Option List
Cognitive/
Gross Motor
Language Skills
For each child, select the best description of development. Sample Nutrition/Biochemistry Option List
For each child with a metabolic abnormality, select the vitamin or mineral that is most likely to be involved. See Appendix B for additional examples.
Chapter 6. Extended-Matching (R-Type) Items
Steps for Organizing a Group to Write Clinical R-Sets
The following steps may be followed to utilize a group in writing clinical R-sets. Some groups have met over dinner; fol-lowed the steps below; and generated a first draft of a dozen or more items from each attendee. Others have scheduled thisas a full-day off-campus retreat, with the goal of generating a pool of near-final items.
The organizer of the “item writing party” should think about how the participant’s time can best be spent. The yield will besubstantially greater in terms of both quality and quantity if some work is done in advance. For example, there will be sig-nificantly smaller yield if the participants are expected to decide what topics to write on; time will be saved if the topics foreach set are defined in advance of the meeting (Step #1 below). Similarly, more items will be developed if a draft of theoptions for each set is developed in advance (Step #3 below). More items will also be developed if a sample item is writtenas a model for each set (Step #4 below). There are situations where it is best to allow item writers considerable flexibilityin determining what to write. In these circumstances, providing one or more option sets and allowing them to generate oneor more options sets on their own might be workable. Decisions will also need to be made about the composition of the pairs of item writers. In some circumstances, it might bebest to allow participants to self-select their partners. In other circumstances, it might be best to assign the pairs. We havehad the most success assembling item-writing pairs with similar interests, but including individuals with more diverse exper-tise as reviewers. If the goal is to generate interdisciplinary items for an exam, the same list of options may be provided toseveral pairs of item writers. For example, options related to abdominal pain could be given to a pair of gynecologists; a pairof surgeons; and a pair of internists. In constructing the exam, one or more items from each discipline could be merged intothe same set, requiring that examinees think across disciplines in determining the most likely diagnoses of the patients.
Significant time can be saved if participants write their items on computers. This saves many hours of trying to decipher thehandwriting of the item writers. We typically have two item writers work at one computer (we have found that someoneautomatically assumes control of the keyboard). Then, several approaches are possible for the review: reviewers may con-gregate around the computer to read the items on screen; the item author can read the item aloud to the reviewers who donot have a copy of the items; or hard copies can be printed for everyone to use in the review session.
1. Define the content domain of the exam. For example, in developing an exam to test the ability to diagnose com-
mon clinical problems, you might define the domain by a list of the chief complaints. Time will be saved if this isdone in advance of the item-writing meeting.
2. Train a group of faculty members to serve as item writers. Training should include a brief discussion of the pur-
pose of the exam, some sample items, and the procedures to be followed during item writing.
3. Divide the group into pairs to write items. Each pair is assigned to write on 2-4 chief complaints; they generate
(or modify) a list of diagnoses for each assigned complaint and write one or more patient descriptions for the diag-noses they included in their option list. Expect 20 to 60 item stems from each pair (10 to 20 per complaint). Use ofcomputers will save considerable time in the long run.
Chapter 6. Extended-Matching (R-Type) Items
4. Stress the following guidelines for writing stems.
Each item should describe a patient with one of the diagnoses in the option list, beginning with the patient’s age, gen-der, chief complaint, and site of care, followed by personal history, family history (if relevant), then physical exam-ination information, then laboratory data (if provided).
Depending upon the purpose of the set, vignettes can be brief prototypic presentations or fuller descriptions that chal-lenge examinees to identify key information.
Each patient description should be similar in structure to the others in the set. For example, if race, ethnicity, oroccupation is included in one item, include it in all items; if laboratory data are included for one item, include themin all items.
5. Merge the pairs into a larger group to review the items. One approach is to have the author read the item aloud;
others attempt to provide the correct answer. The group reviews the option list and modifies the item or the optionlist to eliminate any ambiguity. Other approaches are outlined above.
6. Type, edit, and subject the items to external review. Items should be reviewed without the correct answer indi-
cated after they are in their final form.
7. Construct the test. Select a sample of items from each complaint; save the remaining items for subsequent exams.
Items can be converted into one-best-answer items by adding a lead-in and the best five (or more) options from theoption list.
Form for Writing R-Sets
Theme: _________________________________________________________________(eg, a presenting complaint) Lead-In:_________________________________________________________________(eg, For each patient with fever, select the most likely diagnosis.) Chapter 6. Extended-Matching (R-Type) Items
Comparison of Items in Five-Option and Extended-Matching Format
In several studies to investigate the optimal number of options for multiple-choice items, we have consistently found that,other things being equal, more options are better than fewer options. Based on items used on NBME exams, extended-matching items are more discriminating than any other format; 5-option A-types are second best; and the various forms oftrue/false items are the worst. In controlled studies comparing otherwise equivalent 5-option and extended-matching items,extended-matching items were found to be more discriminating than 5-option items; comparable levels of reproducibility canbe achieved with the extended-matching format using one-third fewer items than with 5-option items. Extended-matchingitems were also found to be more difficult than content-parallel 5-option items: there is a lower probability of guessing thecorrect answer, and item writers are not always able to select the most functional distractors in reducing the number ofoptions to five. The following table shows examinee responses to a sample item presented in a 5-option format and in a 15-option format.
The item was originally written as a 15-option item; the item writer then reduced the number of options to five by selectingwhat he thought were the best distractors (B, F, G, J, N). The item was markedly easier in the 5-option format (p value of81 vs 59), and the discrimination was markedly lower (not shown). There is an increased probability of examinees select-ing the correct answer in the 5-option format, especially because item writers do not uniformly identify the most salient dis-tractors (eg, option D). Tests constructed of extended-matching items tend to spread out the lower ability students; theextended list of options gives them more opportunity to show what they don’t know. Examinee Responses
Case SM, Swanson DB, Ripkey DR. Comparison of items in five-option and extended-matching format for assessment of diagnostic skills. Acad-emic Medicine. 1994;69(suppl):S1-S3.
Case SM, Swanson DB. Extended-matching items: a practical alternative to free-response questions. Teaching and Learning in Medicine.
1993;5(2):107-115.
Swanson DB, Case SM. Trends in written assessment: a strangely biased perspective. In: Harden R, Hart I, Mulholland H, eds. Approaches toAssessment of Clinical Competence. Norwich, England: Page Brothers; 1992:38-53.
A’s to R’s and Back Again
Although research has consistently shown a psychometric advantage of Extended-Matching (R-type) items over five-optionA-type items, there may be circumstances where you need to convert items from one format to another. This should bestraightforward.
For example, the following item was written as a five-option A-type item: A patient with the classic phenotypic features of trisomy 21 (Down syndrome) has 46 chromosomes on each of 100
metaphase karyotypes. Which of the following is the most likely explanation for this finding?
A.
Deletion
*D.
Translocation
B.
Mosaicism
E.
Undetected trisomy
C.
Somatic mutation
This can easily be transformed into the following extended-matching item. Once in this format, additional stems can be writ-ten to expand the set.
A.
Deletion
D.
Translocation
B.
Mosaicism
E.
Undetected trisomy
C.
Somatic mutation
For each patient with genetic abnormalities, select the genetic pattern that is most likely to be involved.
A patient with the classical phenotypic features of trisomy 21 (Down syndrome) has 46 chromosomes on each of 100
metaphase karyotypes.
Additional options could be written to make the examinee task more challenging. A.
Deletion
F.
Somatic mutation
B.
Genomic imprinting
G.
Translocation
C.
Mosaicism
H.
Undetected trisomy
D.
Pleiotropy
I.
Variable expressivity
E.
Reduced penetrance
For each patient with genetic abnormalities, select the genetic pattern that is most likely to be involved.
Items from this set could be converted back to single A-type items with five or more options.
Chapter 6. Extended-Matching (R-Type) Items
Chapter 7
Pick N Items: An Extension of the Extended-Matching Format
The Pick N format may be similar to either the extended-matching or the A-type format; the primary difference is that theexaminee is told to pick 2, 3, 4, or even 5 of the options listed. As with extended-matching sets, the option list may includeup to 26 options. The format was developed to replace negative items or items with double options, particularly in areassuch as Health Maintenance and Disease Prevention. Items might focus on various patients with different risk factors whocome for routine examination; the examinee would be asked to select the laboratory studies or immunizations that should beordered for each patient. The format could also be used for emergency management items where several procedures wouldbe carried out simultaneously; the examinee would be asked to select a specific number of actions from the option list.
The item-writing rules are the same as for extended-matching sets. The options should be short (usually a single word orvery short phrase); the patient vignettes can be long. If the set asks for management decisions, each vignette should containall relevant history and physical examination data. As with extended-matching items, the format works well for items thatappear to be extremely easy; tricky or unnecessarily complex vignettes should be avoided. The Pick N format is designed to specify exactly how many options to select. The rationale for this decision is derived fromthe essential difference between true/false and one-best-answer items, where true/false items require the examinee to indi-cate all responses that are appropriate, and one-best-answer items require the examinee to indicate a specific number ofresponses. Specifying the number of options to be selected changes the task from a true/false task to a best-answer task.
Research indicates that partial credit scoring is preferred but logistical considerations might preclude this. If you use all-or-nothing scoring, the items may be extremely difficult, and it is best to require examinees to select only two or three options,rather than more. Ripkey DR, Case SM, Swanson DB. A “new” item format for assessing aspects of clinical competence. Academic Medicine. 1996;71(suppl):S34-S36.
Chapter 7. Pick N Items: An Extension of the Extended-Matching Format
A 50-year-old man has the gradual onset of mental confusion, disorientation, and loss of short-term memory. He
has a left footdrop. A blood smear shows microcytosis and basophilic stippling of erythrocytes. (Select the two
most likely diagnoses)
In the example above, there would be disagreement about what diagnoses are likely, but the task becomes clear if the exam-inee is told to select the two most likely diagnoses. The options can be diagnosed as follows.
F C E D A G H B
Least Likely Diagnosis
Most Likely Diagnosis
Sample Pick N Set
For each child, select the appropriate vitamin or mineral supplements. 1. A 1-month-old infant is brought to the physician for a well-child examination. He has been exclusively breast-fed, and examination shows normal findings. (SELECT 2 SUPPLEMENTS). 2. A 6-year-old girl has cystic fibrosis. She has been taking no medications. (SELECT 3 SUPPLEMENTS). Ans: E, J, K
Sample Pick N Set
Analysis and culture of cerebrospinal fluid For each child with fever, select the appropriate initial diagnostic studies. 1. A previously healthy 1-year-old girl is brought to the emergency department because of fever for 1 day. Her tem- perature is 41 C (105.8 F). She is otherwise asymptomatic. Physical examination shows no abnormalities.
(SELECT 4 STUDIES).
Ans: B, C, G, I
2. A previously healthy 10-day-old newborn is brought to the emergency department because of fever for 2 hours. He was born at term after an uncomplicated pregnancy. His temperature is 39 C (102.2 F). Physical examination showsno abnormalities. (SELECT 6 STUDIES).
Ans: A, B, C, E, G, I
3. A 7-year-old boy with sickle cell disease is brought to the emergency department because of fever for 1 day and chest pain for 1 hour. His temperature is 39.5 C (103.1 F). Breath sounds are slightly decreased in the right lower lung;he is not in respiratory distress. (SELECT 3 STUDIES).
Ans: B, C, I
Chapter 7. Pick N Items: An Extension of the Extended-Matching Format
PATIENT CHART
Patient History
Current age:
Chief complaint:
Social history:
Marital status:
Occupation:
Alcohol:
Smoking:
Exercise prgm:
Medical history:
Childhood:
Immunizations:
all childhood immunizations; last tetanus toxoid age 15; no immunizations since childhood Screening:
Family history:
Parents:
father age 57; hypertensive mother age 55; obese and hypertensive Siblings:
Children:
Current medications:
Allergies:
Physical Examination
Vital signs:
Blood pressure
Respirations
Abdominal:
Laboratory studies:
1. For the patient whose chart is shown, select the conditions for which he is at increased risk. 2. For the patient whose chart is shown, select the most appropriate health maintenance interventions. Serum urea nitrogen (BUN) and creatinine Chapter 7. Pick N Items: An Extension of the Extended-Matching Format

Source: http://medecine.ulb.ac.be/tools/docimo/NBME_R-Type.pdf