Recruitment of Technical Experts

Experts were sought who could provide content and/or methodological guidance. The group of experts included representatives from the project partners AAD and AAP, as well as from the American Pharmaceutical Association (AphA), Society for Investigational Dermatology (SID), and the Skin Collaborative Review Group (CRG) from the Cochrane Collaboration. Their input was sought through ad hoc correspondence as well as through more formal requests for feedback during the project. Specific requests for feedback were made for key decisions throughout the project, such question refinement, completeness of literature search, treatment and acne severity classification, and format and layout of evidence tables and synthesis of literature.

Additional representation of AAD, AAP, American Academy of Family Physicians (AAFP), as well as from the Society for General Internal Medicine (SGIM), SID, and Society for Adolescent Medicine (SAM) joined the technical experts as peer reviewers for the report. These reviewers included dermatologists, pediatricians, and generalists (Appendix A).

Patient Population

The target population for this report is all patients with acne who do not have complicating comorbidities such as endocrinopathies. In consultation with the partners and technical experts, we decided to exclude evidence from populations with the following conditions: acne fulminans, acne necroticans, acne venenata, rosacea, and chloracne. We believed that to include endocrinopathies would necessitate an examination of underlying medical conditions that was beyond the scope of this report. The search and synthesis of evidence was not limited by age, gender, or any other patient characteristic, though synthesis includes an analysis by these characteristics, where possible.

The report sought evidence concerning all primary care and dermatology settings. In addition, self-treatment is an important modality for acne therapy. Therefore, the community-based setting in which a consumer of health care acts without the involvement of a health care provider was also considered.

Questions

Questions were identified through discussions among the team members, preliminary searches of MEDLINE^®, a search of the MicroMedex pharmaceutical database, and consultation with our technical experts. We also reviewed over 14,000 consumer-generated questions posted to a major consumer health Internet Web site (InteliHealth™) for queries about acne. The questions identified through this process formed the basis of a brief questionnaire distributed to the technical experts (Appendix B). The experts were asked to rate the importance of each question, to provide a sense of the data available to address each question, and also to provide any additional comments or suggestions.

Based on the results from the questionnaire, we developed questions in factorial format to embrace the various patient treatments, outcomes, patient characteristics, practice settings, and care providers of interest (Appendix C). The factorial formatted questions, as well as the proposed exclusions, were distributed to the technical experts for feedback (Appendix D). From the factorial questions, the technical experts requested that the evidence report address the following questions:

1. What treatments are effective in the management of acne?
2. What is the responsiveness of acne to first-line, second-line, and third-line (referral) treatments?
3. What are the side effects of treatment?

The following secondary questions were also to be addressed as the evidence allowed:

• What baseline characteristics raise the likelihood of acne?
• What care do patients provide for themselves at home?
• How effective is initial home care?
• What is the presentation of acne to physicians?
• How compliant are patients with their treatment?
• What is the residuum of untreated acne? Of treated acne?
• What is the quality of life (psychological morbidity) in untreated acne? In treated acne?
• What is the financial cost of therapy?

Causal Pathway

The accumulation of data alone is not sufficient for decisionmaking. The evidence needs to be organized into meaningful units, and the units must be organized to aid in making decisions. To this end, we structured a causal model, using the conventions of an influence diagram ²³ (Figure 1). The diagram is an abstraction of a decision tree, and, when fully specified, can derive the flowchart of a practice guideline. For instance, rather than specifying all possible self-care options, the diagram simply shows that the choice of self-care is influenced by a patient's baseline characteristics which, in turn, influence the possibility and choice of initial medical care. It was developed through structured interviews with two technical experts and a review of the literature, and it was critiqued by the other technical experts.

Figure

Figure 1. Influence diagram of management of acne.

Preliminary definitions and assessments were made of the nodes (states or variables) of the influence diagram:

• Patient characteristics -- These are characteristics such as race, gender, and skin type.
• Self-care -- Refers to over the counter treatments and medications, such as benzoyl peroxide. This also includes self-referral to a medical provider.
• Initial acne assessment -- Refers to the acne present at initial assessment. The arrow from baseline characteristics suggests that this acne severity is influenced by baseline characteristics, although our initial review of the evidence suggested that few investigators go beyond lesion count and type in making a severity assessment (see Chapter 3).
• Initial medical care -- Represents all possible prescription and non-prescription treatments.
• Acne reassessment -- The same as initial acne assessment, but later in time. An arrow is drawn from the initial acne assessment node, suggesting that response to treatment (from initial medical care) is influenced by the initial acne status (or baseline characteristics).
• Initial compliance -- Acknowledges that success depends on patient compliance and that initial medical choices must be made keeping compliance in mind.
• Followup medical care -- Refers to the provider's options after assessment, including continuing present therapy, changing to another treatment, or referral.
• Followup compliance -- Similar to initial compliance.
• Followup acne reassessment -- Similar to acne reassessment.
• Further medical care -- Focuses on referral if acne reassessment shows no improvement. Refers to the provider's option after assessment, including continuing present therapy, changing to another treatment, or referral. We left the full definition open, hoping to find evidence in the literature that would help us to define the therapeutic choices.
• Acne residual -- This is the long-term dermatological morbidity of acne, with or without treatment.
• Quality of life -- Outcomes such as self-esteem and social and familial relationships.
• Cost -- Includes costs related to loss of productivity in school or work, as well as medical costs.
• Other morbidity -- Other morbidity outcomes, such as the presence of lesions themselves, and the possible effects of treatment. Possible treatment effects include sun sensitivity and, of the most concern, more serious side effects such as those related to isotretinoin use.

Influence diagrams terminate in a utility node that integrates the outcomes, their probabilities, and the decisionmaker's valuation of preference for those outcomes. We plan to leave that integration to the users of our final evidence synthesis for creating their guidelines.

Literature Search Methods

Searching the literature included the steps of identifying reference sources, formulating a search strategy for each source, and executing and documenting each search.