Measuring Consequences in Patients with ACL Rupture
Summary af Jonathan Comins
The objective of this PhD study was to construct and validate a patient-related outcome (PRO) questionnaire that can be used to measure the effect of treatment in patients with ACL deficiency. This process was divided into separate studies described in the three articles presented in this thesis.
Study 1 encompassed a literature search to find all questionnaires used to assess outcome in the targeted patient group. The objective was to identify item content deemed suitable for these patients by clinical experts; that is, items that possess face validity. The next step was to translate all “non-Danish” items into Danish and consolidate items with redundant content. Different questionnaires ask many of the same questions. Thus, item reduction was performed retaining only items with unique content.
The literature search included 31 PROs, which yielded 539 items in four languages. Because the majority of items were not in Danish, translation was carried out by extracting just the meaningful content of the item. For example, an item such as “In the past week, I have had difficulty walking down a flight of stairs” would be truncated to “difficulty walking down stairs”. These truncated items were then translated to Danish and assessed for content redundancy. The final number of truncated items with unique content was 157.
This process was the substantive part of Paper 1. The article was submitted for publication and is now under revision.
The second study involved focus group interviews with patients with ACL rupture, pre- and postoperatively. Each item from the literature search was discussed on an item-by-item basis to ascertain the content relevance for these patients. Thirty-eight items from the initial item pool, five modified items, and twelve items with new content were confirmed to be relevant by the patients in three focus groups and seven single interviews.
The result was a 55-item pilot questionnaire with six proposed functional measurement domains. This process is described in Paper 2, which has been submitted for publication and is in review.
In the third study, 242 patients consisting of patients prior to and subsequent to ACL reconstruction were recruited from the ACL registry list at Bispebjerg Hospital. The patients completed the 55-item draft questionnaire. The subjects consisted of three groups: 62 subjects in the pre-operative group, 87 subjects in the first post-operative group (four to 16 months post-op), and 93 subjects in the third post-operative group (at least 28 months post-operative).
The responses were analyzed using Partial-Credit and Graphical Loglinear Rasch models. Forty-one items exhibited fit to the Rasch model and thus possess unidimensional measurement characteristics as applied to patients, pre- and post- ACL reconstruction. The items were
distributed across seven constructs and not six as proposed a priori. This was because one
proposed domain was found to consist of two separate constructs.
The seven scales comprise the newly formed condition-specific PRO questionnaire entitled the Knee Numeric Entity Evaluation Score – ACL (KNEES-ACL).2 This scale validation study is described in Article 3, which is included in this thesis.