Developing a model

To develop a model of shared care appropriate for our Service, we used a modified Delphi method, acknowledging that the components or elements of treatment for opioid dependence are largely either regulated or well established. We used this method to explore general elements of shared care, patient stability, required assessments, approved prescriber responsibilities and psychosocial support.

The Delphi method is a structured communication relying upon experts to make decisions. They answer questionnaires in two or more rounds and after each round are provided an anonymised summary of the decisions from the previous round including any reasons provided for their decisions. In subsequent questionnaires, the experts may revise their earlier answers in light of the previous questionnaire results. It is expected that the range of the answers will decrease in this process, and the group will converge towards the "correct" decision. Finally, the process is completed once a predefined criterion is reached. You can decide the criterion – for example, your Service may decide that 80% of respondents provides consensus. So if 80% of respondents make the same decision, you have consensus on that decision. The decision may need further refinement if respondents have provided further comments for consideration.

Our first questionnaire then, rather than asking open questions, was created incorporating components of shared care models used within Queensland, NSW and internationally. Clinicians were presented with clinical guideline statements and asked to provide yes/no answers in response to the statements and provide further commentary if they chose.

The responses were used to refine the model and generate other elements for consideration.

The second questionnaire asked the clinicians to reconsider some elements, refine others and consider new elements. 

And finally, a third questionnaire was conducted as a face-to-face discussion with medical and nurse practitioners and nursing team leaders. It was conducted this way as the elements that needed clarification were mostly related to prescribing and service structure issues.

This method provides an opportunity for all clinical staff members to collaborate in developing the model and therefore hopefully, taking ownership of the model.

Here you will find copies of the three questionnaires, provided as examples that you could use to guide your considerations. Remember; the second and third  questionnaires are based on the responses from the previous questionnaire/s, meaning that if you were to use the first questionnaire for your service, the second might look very different from this example – dependent on your expert panel responses.