JMIR Publications

ABSTRACT

Background:

Musculoskeletal symptoms, such as neck and shoulder pain and stiffness and low back pain, are common health problems in the working population. They are the leading causes of presenteeism (employees being physically present at work but unable to be fully engaged). However, current medical systems do not spare sufficient resources for non-specific musculoskeletal problems.

Objective:

This study aimed to evaluate the improvements in musculoskeletal symptoms after use of an exercise-based artificial intelligence (AI)-assisted interactive health promotion system that operates through a mobile messaging app (the AI-assisted health program).

Methods:

We conducted a two-armed, randomized, controlled, and unblinded trial in workers with neck/shoulder stiffness and/or low back pain. We recruited participants with these symptoms through email notifications. We obtained 48 participants in the intervention group and 46 in the control group. The intervention group received the AI-assisted health program, in which the chatbot sent messages to users with the exercise instructions at a fixed time every day through the smart phone’s chatting app (LINE) for 12 weeks. The exercises could be performed within 1 minute. The control group continued with their usual care routines, which included exercising for 3 minutes at recess time provided by the company to prevent stiff shoulders and back pain. We assessed the subjective severities of the neck and shoulder pain/stiffness and low back pain in participants using a scoring scale of 1 to 5 for both the intervention and the control group at baseline and after 12 weeks of intervention using an online form.

Results:

We analyzed 47 patients in the intervention group and 40 in the control group. The participants in the intervention group showed significant improvements in the severities of the neck/shoulder pain/stiffness and low back pain compared to those in the control group (OR 12.74, P <.001). Based on the subjective assessment of the improvement of the pain/stiffness at 12 weeks, 36 (77%) participants in the intervention group and 3 (8%) in the control group had improved (improved, slightly improved) (OR 54.23, P <.001).

Conclusions:

This study showed that the short exercises provided by the AI-assisted health program improved both neck/shoulder pain/stiffness and low back pain in 12 weeks. Digital health programs are low cost and safe and can save experts’ working hours and labor costs. Further studies are needed to identify the elements of the AI-assisted health program that worked. Clinical Trial: University hospital Medical Information Network-Clinical Trials Registry (UMIN-CTR) 000033894; https://upload.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000038307.