Table 5.

Type, features, and outcomes measured for the CDSS^a for treatment recommendation.

Study and CDSS name	CDSS type	Features of the CDSS	Outcomes measured	Results
Benditz et al [38], 2019	Knowledge based	Questionnaire-based CDSS A computerized tool with disease-specific algorithms cascading the next best questions leading to the most probable diagnosis and actions	Association between the diagnoses and treatment recommendation of the tool and the physician’s diagnosis	Significant correlation with small-to-medium effect between the DSSb and the medical recommendation. Cramer V=0.293; P=.02 Concordance: 49.6% Overestimated: 36% Underestimated: 14.4%
Byvaltsev and Kalinin [55], 2021	Nonknowledge based	Semantic network structured based on the medical ontology and fuzzy logic principles Computer-assisted electronic checklist and recommendations, which uses preoperative instrumental data on lumbar segments of patients with degenerative diseases	Pain using visual analog scale Lower limbs Lumbar spine ODIc	For patients who underwent total disc replacement, pain syndrome level and functional status were comparable before surgery, on discharge and 3 mo after surgery (P>.05). A total of 6 mo after the surgery, there was a decrease in pain intensity in the lower limbs (P=.02) and lumbar spine (P=.03) and an increase in functional status by ODI (P=.02) in the CDSS group. In the CDSS patients group who underwent minimally invasive rigid stabilization, there was a decrease in pain intensity in the lower limbs (P=.01 for both 3 mo and 6 mo after surgery) and in the lumbar spine (P=.03 and P=.02 for 3 mo and 6 mo after surgery, respectively) and an increase in functional status by ODI (P=.01 and P=.03 for 3 mo and 6 mo after surgery, respectively). For patients who underwent open rigid stabilization, pain syndrome level and functional status were comparable (P>.05) before surgery, on discharge and 3 mo after the surgery. A total of 6 mo after surgery, there was a decrease in pain intensity in the lower limbs (P=.04) and lumbar spine (P=.03) and an increase in functional status by ODI (P=.01) in the group using CDSS.
Downie et al [56], 2020	Knowledge based	Decision tree app–based CDSS It consists of a knowledge base, reasoning engine, and interface. An advice report will be generated after the history and screening inputs. The pharmacist may add any key message or modify the advice.	Qualitative-based CDSS: Ease of use, consistency (of visual language or interaction model), system visibility, navigation or workflow, content, clarity, and acceptance System usability scale Level of acceptance of clinical reasoning and decision support	Ease of use: mostly negative sentiments (16/26, 62%) Consistency: mostly neutral sentiments (7/13, 54%) Visibility: mostly negative sentiments (7/16, 44%) Navigation or workflow: mostly neutral sentiments (12/16, 75%) Content: mostly positive sentiments (12/27, 44%) Clarity: mostly neutral sentiments (9/15, 60%) Acceptance: mostly positive sentiments (34/49, 69%) System usability scale Overall system usability: excellent (mean 0.92, SD 6.5), with acceptance rated as good to excellent. CDSS-pharmacists' agreement: Self-care recommendations: 90% (18/20) Medicines recommendations: 100% (25/25) Referral advice :88% (22/25) Pharmacists expressed uncertainty when screening for serious pathology in 40% (10/25) of the cases. Pharmacists requested more direction from the CDSS in relation to automated prompts for user input and page navigation.
Back-UP (Horizon 2020), Jansen-Kosterink et al [57], 2021	Knowledge based	Binary logistic regression Short questionnaires were completed by patients that stratified them into 1 of the 3 outcome groups. Targeted interventions were recommended for each outcome group.	Qualitative-based CDSS: Factors that promote or hinder the acceptance of clinicians toward CDSS use	Reason to use a complex CDSS: Improve care of patients (assessment, n=20) Curiosity to test and use the CDSS, to see for themselves what the value of the system is (n=19) Expectation of increase in efficiency because of reduction of workload and time and allowing them to reorganize work (n=18) Support during decision-making (n=17) Patient empowerment (n=14) Work consistently with evidence-based medicine (n=8) Perceived the technology as friendly to use (n=3) Barriers to using a complex CDSS: Worried about their own clinical practice and autonomy; physicians are reluctant to use a CDSS when it interferes too much with clinical practice (n=18) Do not want to use a CDSS when it comes at an increase in time and costs (n=18) The fear that the CDSS does not work correctly (n=17) A too generic approach (n=15) A lack of effectiveness and added value (n=11) Hampering personal contact with the patient (n=8) Data and security issues (n=8) Capitalizing on health care (n=4) Lack of trust (n=3) If the use of CDSS is imposed by external parties, such as health care insurance companies (n=3)
SLICd CDSS (Kubben), Kubben et al [58], 2011	Knowledge based	Decision tree app–based CDSS Offers evidence-based algorithms (eg, burst fractures, central cord syndrome, facet fracture dislocation, facet subluxation, and hyperextension injury) based on the Subaxial Injury Classification system	Not reported	Not reported
Peiris et al [41], 2014	Knowledge based	Recommendations from 15 guidelines for back pain management After excluding serious pathology, the CDSS will continue to assess for the most probable diagnosis and treatment through a series of questions. A personalized information sheet will be printed.	Frequency of use of the web-based tool by physicians Acceptability by physicians	Acceptability by physicians Considered that back pain is easy to manage and the use of CDSS could insult their skills Found CDSS useful for patient reassurance and minimizing complex tests or treatment. Suggestions for improvement: Increase comprehensiveness of advice for complex pain management and referral and allow customization of advice Integration of software systems and easy navigation

^aCDSS: clinical decision support system.

^bDSS: decision support system.

^cODI: Oswestry Disability Index.

^dSLIC: Subaxial Injury Classification.