Chronic Cough In detail

Guideline Segment I:

Chronic cough is cough that lasts for at least 3 weeks. Chest radiographs should be ordered before any treatment is prescribed in nearly all patients with chronic cough (Grade II-2). Chest radiographs do not have to be routinely obtained before beginning treatment for presumed PNDS (post nasal drip syndrome) in young nonsmokers, in pregnant women, or before observing the result of discontinuation of an ACE-I (ACE Inhibitor) for 4 weeks for patients who developed cough shortly after beginning to take an ACE-I.

1. The Patient information enquiry is the first task in the workflow. It collects the following data items:
   1. Patient’s age group (Younger/Older)
   2. Patient’s gender
   3. Cough duration
   4. Is the patient a smoker?

2. The Guideline is not appropriate action has a precondition stating that it would only run if cough duration is less than 3. Cough duration of less than 3 weeks is not considered chronic and therefore in such cases the guideline is ineligible.

3. If the cough is chronic, a follow-up enquiry is invoked. Depending on the gender of the patient, one of the Initial assessment enquiries will run while the other will be discarded. The two enquiries are identical, except that the Initial assessment - female patient enquiry collects one extra data item, which is whether the patient is pregnant or not. The other data items collected by these enquiries are:
   1. Has the cough developed shortly after the patient started taking an ACE-I?
   2. Post nasal drip syndrome?
   3. Is the cough productive? (information used in a following guideline segment)

4. The CXR and initial treatment plan contains a decision and two plans.

1. The Scheduling decision determines whether a chest radiograph should be ordered before initial treatment. Pregnancy, PNDS in young non-smokers and ACEI-related cough are all arguments in favour of beginning treatment before ordering chest radiographs (the CXR_in_parallel candidate). In all other cases, a chest radiograph should be ordered before any treatment is prescribed; thus, the argument for the CXR_first candidate is that there are no arguments for the CXR_in_parallel candidate.

2. The CXR first plan has a precondition stating that it would only run if the result of the Scheduling decision is CXR_first (otherwise, it would be discarded).



The first task in the CXR first plan is to order a chest radiograph.



Next, the result of the radiograph is requested (Normal/Other).



Finally, appropriate treatment is recommended (this is out of the scope of the current model).



3. The CXR in parallel plan has a precondition stating that it would only run if the result of the Scheduling decision is CXR_in_parallel (otherwise, it would be discarded).



The Order CXR action and the Initial treatment of chronic cough plan run in parallel.



The Initial treatment of chronic cough plan contains two actions. The Non-specific initial treatment action recommends appropriate non-specific treatment (this is out of the scope of the current model); the Discontinue ACEI action has a precondition stating that it would only run if the cough is ACEI-related, in which case the clinician should observe the result of discontinuing the ACEI for 4 weeks.

Guideline Segment II:

When the chest X-ray is normal, PNDS, Asthma, and GERD (Gastroesophageal reflux disease) are the likely causes of chronic cough.

5. The Investigations plan further investigates the cause of the cough.

1. The Cause decision has 4 candidates: PNDS, GERD, Asthma and Other. A normal chest radiograph result is an argument for PNDS, GERD and Asthma. A result that isn’t normal is an argument for a different cause (Other). The radiograph result is collected by the CXR_result data item in one of the CXR report enquiries.

As several candidates could be supported by the same argument, the Cause decision allows for multiple-selection of candidates.

Guideline Segment III:

In PNDS, sinusitis may be the cause up to approximately 30% of the time when cough is nonproductive, and up to approximately 60% of the time when cough is productive. Sinus CT scans are not routinely recommended to evaluate for sinusitis as the cause of cough. Four-view sinus radiographs should be ordered instead.

6. The PNDS evaluation plan follows the Cause decision. It has a precondition stating that it would only run if the results of the Cause decision include the PNDS candidate.

1. The PNDS evaluation plan contains two Sinusitis actions. One would run if the cough is productive, and the other would run if it isn’t (this information is collected by the Cough_productive data item in the one of the initial assessment enquiries). The two actions carry information about the probability of sinusitis being the cause of cough, and a negative recommendation about using CT scans to evaluate for sinusitis.

2. The sinusitis action is followed by the Sinus radiographs action, which recommends 4 view sinus radiographs.

Guideline Segment IV:

While 24 hour esophageal pH monitoring is the most diagnostically useful test for assessing for GERD as the cause of cough, conventional indices used by gastroenterologists to assess for esophagitis may be misleadingly normal. Therefore, until future studies provide better guidelines, the test should be read as normal when conventional indices are within the normal range and no suspicious refluxinduced coughs appear during the monitoring session (Grade II-2).

7. The GERD evaluation plan follows the Cause decision. It has a precondition stating that it would only run if the results of the Cause decision include the GERD candidate.

1. The first task in the GERD evaluation plan is the Oesophageal pH monitoring action, as this is the most diagnostically useful test for assessing for GERD as the cause of cough.

2. Next, the Oesophageal pH monitoring results enquiry collects two data items:
   • Results according to conventional indices (Within normal range / Outside normal range)
   • Suspicious coughs appearing in the monitoring session (Yes / No)

3. Finally, the Oesophageal pH monitoring interpretation decision determines whether the test should be read as normal. It has two candidates, Normal and Abnormal.
   • The Normal candidate is supported by an argument stating that the conventional indices are within the normal range and that there were no suspicious coughs during the monitoring session
   • The Abnormal candidate has no supporting arguments. Instead, it has a modified decision rule stating that it would be recommended if there is no support for the Normal candidate.

 

If you have installed the Tallis Toolset, you can find this sample in the World > Samples folder in the ACL Web Repository.

Run this sample file

Click her for information about using the web enactment application.

This model is based on a model that was created as part of a study comparing six computer-interpretable guideline models (Peleg M, Tu S, Bury J, Ciccarese P, Fox J, Greenes RA, Hall R, Johnson PD, Jones N, Kumar A, Miksch S, Quaglini S, Seyfang A, Shortliffe EH, Stefanelli M. Comparing computer-interpretable guideline models: a case-study approach. J Am Med Inform Assoc. 2003 Jan-Feb;10(1):52-68.)