Skip to main content

Advertisement

Table 3 Suitability of workflow control patterns for CPGs

From: Leveraging workflow control patterns in the domain of clinical practice guidelines

Pattern Suitable Explanation Example
1. Sequence yes - Stabilize on tocolytics before transfer mother to appropriate level of care if possible [ 27 ]
2. Parallel split (AND split) yes - The maternal pulse should be felt simultaneously to differentiate between maternal and fetal heart rate [ 27 ]
3. Synchronization (AND join) yes explicit/implicit in a sequencing after a parallel split Zidovudin for one hour and single dose of Nevirapine 30 min before the skin incision. Afterwards give Retrovir until birth [ 27 ].
4. Exclusive choice (XOR split) yes - If diastolic blood pressure >140 mmHg occurs on two readings 5 min apart, then start a continuous IV infusion of an antihypertensive agent [ 28 ]
5. Simple merge (XOR join) yes explicit/implicit in a sequencing after an exclusive choice … locoregional postoperative radiotherapy (after BCT or MRM) [ 29 ]
6. Multi-choice (OR split) yes - Add regular treatment with one or more bronchodilators [ 30 ].
7. Structured synchronizing merge (OR join) yes explicit/implicit in a sequencing after a multi-choice (see example above)
8. Multi-merge no multiple activation of a task only in structured loops  
9. Structured discriminator unknown synchronization of 2 or more branches waiting for the first incoming branch has not been found in guidelines  
10. Arbitrary cycles no multiple activation of a task only in structured loops  
11. Implicit termination yes by definition  
12. MIs without synchronization no no multiple, concurrent instances of a task  
13. MIs with a priori design-time knowledge no
14. MIs with a priori run-time knowledge no
15. MIs without a priori run-time knowledge no
16. Deferred choice yes - Surgery to reduce tumour load. It is unclear whether limited or radical surgery is better [ 29 ] .
17. Interleaved parallel routing unknown WCP-40 with the possibility of adding partial ordering constraints has not been found in sample guidelines  
18. Milestone (deadline) yes WITH or WITHOUT activity disablement beyond the milestone In patients hypoxemic during a COPD exacerbation, arterial blood gases and/or pulse oximetry should be evaluated prior to hospital discharge (WITH) [ 30 ] ICD implantation is reasonable in selected patients with LVEF < 30-35 %, not within 40 days of a myocardial infarction, on optimal background therapy … (WITHOUT) [ 31 ]
19. Cancel activity yes   If renal function deteriorates substantially, stop treatment [ 31 ]
20. Cancel case yes - Patients with apparent exacerbations of COPD that do not respond to treatment should be re-evaluated for other medical conditions [ 30 ]
21. Structured loop yes typically in follow-up activities Follow-up: first year: once every three months; second year: once every six months; subsequently: annually [ 29 ].
22. Recursion no multiple activation of a task only in structured loops  
23. Transient trigger yes interpreted as triggers to be acted on immediately Administer controlled oxygen therapy and repeat arterial blood gas measurement after 30–60 min [ 30 ].
24. Persistent trigger yes interpreted as triggers to be acted on either immediately or at some future time Spirometry should be performed if there is a substantial increase in symptoms or a complication [ 30 ] .
25. Cancel region yes - Discontinue drugs that may lower heart rate in presence of bradycardia [ 31 ].
26. Cancel MI activity no no multiple, concurrent instances of a task  
27. Complete MI activity no
28. Blocking discriminator no
29. Cancelling discriminator no no concurrent execution of tasks within cancelation  
30. Structured partial join unknown synchronization of 2 or more branches waiting for the first N incoming ones has not been found in guidelines  
31. Blocking partial join no no multiple execution threads  
32. Cancelling partial join no no concurrent execution of tasks with cancelation  
33. Generalized AND join no no multiple execution threads  
34. Static partial join for MIs no no multiple, concurrent instances of a task  
35. Cancelling partial join for MIs no
36. Dynamic partial join for MIs no
37. Acyclic synchronizing merge no useful for non-structured models only  
38. General synchronizing merge no multiple activation of a task only in structured loops  
39. Critical section yes - The simultaneous administration of radiotherapy and chemotherapy … is discouraged [ 29 ]
40. Interleaved routing yes - Based on ‘expected survival benefit’ no statement can be made. Regarding the optimum sequence of radiotherapy and chemotherapy … The simultaneous administration of radiotherapy and chemotherapy … is discouraged [ 29 ]
41. Thread merge no no multiple execution threads  
42. Thread split no
43. Explicit termination yes - Discharge with planned follow-up [ 32 ]
  1. Legend of ‘Suitable’ column: ‘yes’ indicates that the pattern has been found in sample CPGs; ‘unknown’ that it has not been found in sample CPGs; and ‘no’ that there exist strong reasons to dismiss the applicability of the pattern in the CPG domain. Bold rows present the workflow control patterns that were finally deemed suitable