Every professional discipline today is being called upon to document its effectiveness. Current questions often asked of professionals are:

• "How do we know if current professional practices are effective in bringing about the desired outcomes?"
• "Are there other approaches, or modifications of existing approaches, that might produce better outcomes or similar outcomes at less cost?"

The difficulty in answering these questions is that many times the methods used in current professional practice have not been studied extensively or rigorously.

Evidence-based clinical practice guidelines are intended to help professionals, parents, and others learn what scientific evidence exists about the effectiveness of specific clinical methods. This information can be used as the basis for informed decisions. This guideline represents the panel's concerted attempt to interpret the available scientific evidence in a systematic and unbiased fashion and to use this as the basis for developing guideline recommendations. It is hoped that, by using this process, the guideline provides a set of recommendations that reflects current best practices and will lead to optimal outcomes for children with developmental problems.

Strength of Evidence Ratings for Guideline Recommendations
Each of the guideline recommendations in Chapters III and IV is followed by a "strength of evidence" rating designated by the letter [A], [B], [C], [D1], or [D2] in brackets immediately after the recommendation. These strength of evidence ratings indicate the amount, general quality, and clinical applicability (to the guideline topic) of scientific evidence used as the basis for each guideline recommendation.

Evidence Ratings

The strength of evidence rating does not indicate whether the recommendation is for or against use of a method. For example:

• If there was strong evidence that an intervention is effective, a recommendation for use of the method would have an [A] evidence rating.
• If there was strong evidence that an intervention is not effective, recommendation against use of the method would also have an [A] evidence rating.
• If a systematic literature review found no evidence about the efficacy of a specific clinical method, the recommendation either for or against use of this method would have a [D1] evidence rating to indicate this was based on panel opinion.
• If a systematic literature review was not done for a specific method, the recommendation either for or against the use of that method would have a evidence rating to indicate this was based on panel opinion.

In developing evidence-based clinical practice guidelines, the process of reviewing the scientific literature to find evidence-based answers to specific clinical questions is challenging. Many times the specific clinical issue of interest may not have been studied extensively in well-designed studies of the type that can adequately determine if a clinical method is effective. At other times, even when well-designed studies have been done on a particular clinical topic, the study findings themselves seldom present totally straightforward and unambiguous answers to the clinical questions of interest

Careful analysis of the studies and considerable judgment are always needed when using the findings of research studies to help in making informed clinical decisions and developing clinical practice guidelines. For most clinical topics, it is exceptional to find studies that evaluate exactly the clinical situations and types of subjects that are of interest. Therefore, it is almost always necessary to generalize to some extent in terms of the subject characteristics (such as age or IQ) and the clinical setting or the type of assessment or intervention method used. The research studies reviewed for this guideline were no exception.

In using research evidence to help make clinical decisions, the two primary considerations are the quality of the evidence and its clinical applicability. to the question of interest.

Primary Considerations in Using Research Evidence

• The quality of the study is primarily related to the study design and controls for bias; the higher the quality of the study, the more confidence there is that the findings of the study are valid. Confidence in the study findings becomes even greater when multiple well-designed studies done by independent researchers find similar results.
• The clinical applicability of a study is the extent to which the study's results would also be expected to occur in the particular clinical situation of interest. The applicability of a study's findings is considered to be greater when the subject characteristics, clinical methods, and clinical setting are similar to the study and clinical situation of interest.

The overall usefulness of a study's findings to clinical decision-making relates both to confidence in the results (based on the quality and amount of scientific evidence) and the similarity of the study's subjects, clinical methods, and setting to the question of interest (that is, its applicability).

Considerations about quality of studies

The considerations about using scientific evidence as the basis for clinical decisions apply to all the recommendations in this guideline. For some of the clinical questions of interest, several studies were found that met criteria for adequate evidence about efficacy. For other questions of interest, few or no studies were found that met such criteria.

There are also numerous articles in the scientific literature that did not meet criteria for adequate evidence about efficacy, yet still contained valuable information that may be useful in clinical practice. These articles include case reports, case series (sometimes using pre and post-test designs), and other descriptive studies, as well as articles that primarily discuss theory or opinion.

Relatively rigorous criteria were used for selecting studies that would provide adequate evidence about efficacy. The findings of these studies were used as the primary basis for developing guideline recommendations. In many cases, information from the articles and studies not meeting these evidence criteria was also reviewed by the panel, but information from these sources was not considered evidence about efficacy and was not given as much weight in making guideline recommendations.

Considerations about applicability of studies

Of particular concern for this guideline was finding high-quality scientific studies that focused on children under the age of 3 years. For some topics, studies were found that evaluated only children within the guideline's target population (children from birth to 3 years of age), but for other topics, the only studies found evaluated groups that included somewhat older children (over age 3).

As noted above, the inclusion of children over age 3 does not affect the quality of the study or bias the results, but it may make the study's findings somewhat less applicable to the guideline topic. Ratings of applicability for this guideline are described in Appendix A, Table A-6.

Applicability was taken into account when making guideline recommendations and more weight was given to findings from high-quality studies that focused on children under 3 years old. However, when there were few good studies found that focused on children in the target age group, then the panel thought it important to generalize from evidence found in good studies of somewhat older children.

Judging the quality and applicability of the evidence when making guideline recommendations

Given the considerations above, the panel needed to carefully evaluate the quality and applicability of the scientific evidence that was used as the basis for these guideline recommendations. Similar limitations and considerations apply to all evidence-based practice guidelines. The strength of evidence ratings are a reflection of both the amount and quality of the scientific evidence found and its applicability to the guideline topic.

http://www.health.state.ny.us/community/infants_children/early_intervention/disorders/autism/chapter1.htm#practice