Decision theoretical foundations of clinical practice guidelines: an extension of the ASH thrombophilia guidelines

Abstract

Decision analysis can play an essential role in informing practice guidelines. The American Society of Hematology (ASH) thrombophilia guidelines have made a significant step forward in demonstrating how decision modeling integrated within Grading of Recommendations Assessment, Developing, and Evaluation (GRADE) methodology can advance the field of guideline development. Although the ASH model was transparent and understandable, it does, however, suffer from certain limitations that may have generated potentially wrong recommendations. That is, the panel considered 2 models separately: after 3 to 6 months of index venous thromboembolism (VTE), the panel compared thrombophilia testing (A) vs discontinuing anticoagulants (B) and testing (A) vs recommending indefinite anticoagulation to all patients (C), instead of considering all relevant options simultaneously (A vs B vs C). Our study aimed to avoid what we refer to as the omitted choice bias by integrating 2 ASH models into a single unifying threshold decision model. We analyzed 6 ASH panel's recommendations related to the testing for thrombophilia in settings of "provoked" vs "unprovoked" VTE and low vs high bleeding risk (total 12 recommendations). Our model disagreed with the ASH guideline panels' recommendations in 4 of the 12 recommendations we considered. Considering all 3 options simultaneously, our model provided results that would have produced sounder recommendations for patient care. By revisiting the ASH guidelines methodology, we have not only improved the recommendations for thrombophilia but also provided a method that can be easily applied to other clinical problems and promises to improve the current guidelines' methodology.

Graphical abstract

Figure 1.

ASH two choices model vs a 3-choices decision model. (A) ASH modeling approach for determining the effect of thrombophilia testing. The model starts with the population considered for thrombophilia testing. Thrombophilia testing refers to testing for any type of thrombophilia or a specific type. Intervention is the course of action other than usual care. Depending on the particular question, this means prescribing thromboprophylaxis, withholding thromboprophylaxis, extending thromboprophylaxis, stopping thromboprophylaxis, withholding birth control pills, or withholding hormone replacement therapy. Usual care typically consists of short-term (3-6 months) anticoagulation (provoked VTE) or indefinite treatment (unprovoked VTE). P-thrombophilia prevalence (denoted in the manuscript as T_p); incidence risks of VTE recurrence is denoted in the manuscript as p_t+ and p_t- for patients with (thrombophilia) positive results and for patients with negative test results, respectively; Association refers to RR for recurrent VTE in patients with thrombophilia vs patients without thrombophilia (RR_t); Relative effects of intervention (anticoagulant) on VTE recurrence (RR_rx) and bleeding (RR_bleed) compared with no intervention. (B) A decision tree showing a 3-choice clinical dilemma: administer treatment (anticoagulants) vs performing a diagnostic test (T) (thrombophilia testing) vs withholding therapy. Each treatment consists of the management strategies “treat all patients,” “treat none,” and “use thrombophilia test” to decide whether to treat. By “treatment,” we refer to a commitment to a course of action that may include management consisting of treatment or diagnostic testing. p_t+=Pr(D+|T+) refers to the probability of VTE recurrence when the thrombophilia test is positive (T+). U1 to U4, utilities (outcomes; see Appendix 1 for details).

Figure 2.

Number of VTE and major bleeding (low bleeding risk scenario). The impact analysis displaying the total number of VTE and major bleeding incurred for ASH panel recommendation R1 (A), R2 (B), R3 to R5 (C), and R6 (D) in the low–bleeding risk scenario. Five decision strategies are shown (from left to right): treat according to the threshold (Rx threshold; equation 3 in the manuscript); perform testing and act accordingly; test according to thresholds (equation 2 in the manuscript); treat none (give anticoagulants to no patient without testing); and treat all (provide anticoagulants for all patients without testing).

Figure 3.

Number of VTE and major bleeding (high bleeding risk scenario). The impact analysis displaying the total number of VTE and major bleeding incurred for ASH panel recommendations R1 (A), R2 (B), R3 to R5 (C), and R6 (D) in the high–bleeding risk scenario. Five decision strategies are shown (from left to right): treat according to the threshold (Rx threshold; equation 3 in the manuscript); perform testing and act accordingly; test according to thresholds (equation 2 in the manuscript); treat none (give anticoagulants to no patient without testing); and treat all (provide anticoagulants for all patients without testing).

Figure 4.

Threshold decision model analysis. The results of the threshold decision model analysis in the setting of the low bleeding risk (A) and high bleeding risk (B). The vertical lines (ASH R∗) refer to the recommendations 1 to 6 by the ASH thrombophilia panel. Theoretical thresholds above or below which treatment vs thrombophilia testing vs no anticoagulant treatment should be given are denoted by P_tt, P_t, and P_rx, respectively (see equations (1), (2), (3)). Note that because all ASH R lines are to the right side of P_rx that is, larger than the treatment threshold P_rx in a low-risk bleeding scenario, offering indefinite anticoagulant treatment to all patients represents the best management strategy (A). The same holds for ASH R1, R3 to R5, and R6 in the setting of high-risk bleeding. Because the vertical line ASH R2 is to the left, that is, lower than P_tt (test–no treatment threshold), discontinuing anticoagulation after 3 months of treatment after VTE due to surgery is recommended (see manuscript for details, Table 3, R2).