Extended mobility scale (AMEXO) for assessing mobilization and setting goals after gastrointestinal and oncological surgery: a before-after study

Background: Early structured mobilization has become a key element of Enhanced Recovery After Surgery pro‑ grams to improve patient outcomes and decrease length of hospital stay. With the intention to assess and improve early mobilization levels, the 8-point ordinal John Hopkins Highest Level of Mobility (JH-HLM) scale was implemented at two gastrointestinal and oncological surgery wards in the Netherlands. After the implementation, however, health‑ care professionals perceived a ceiling effect in assessing mobilization after gastrointestinal and oncological surgery. This study aimed to quantify this perceived ceiling effect, and aimed to determine if extending the JH-HLM scale with four additional response categories into the AMsterdam UMC EXtension of the JOhn HOpkins Highest Level of mObil‑ ity (AMEXO) scale reduced this ceiling effect. Methods: All patients who underwent gastrointestinal and oncological surgery and had a mobility score on the first postoperative day before (July–December 2018) or after (July–December 2019) extending the JH-HLM into the AMEXO scale were included. The primary outcome was the before-after difference in the percentage of ceiling effects on the first three postoperative days. Furthermore, the before-after changes and distributions in mobility scores were evaluated. Univariable and multivariable logistic regression analysis were used to assess these differences. Results: Overall, 373 patients were included (JH-HLM n=135; AMEXO n=238). On the first postoperative day, 61 (45.2%) patients scored the highest possible mobility score before extending the JH-HLM into the AMEXO as com‑ pared to 4 (1.7%) patients after (OR=0.021, CI=0.007–0.059, p<0.001). During the first three postoperative days, 118 (87.4%) patients scored the highest possible mobility score before compared to 40 (16.8%) patients after (OR=0.028, CI=0.013–0.060, p<0.001). A change in mobility was observed in 88 (65.2%) patients before as compared to 225 (94.5%) patients after (OR=9.101, CI=4.046–20.476, p<0.001). Of these 225 patients, the four additional response categories were used in 165 (73.3%) patients