To develop and examine an interactive, multi-functional mobile app-based technology designed to encourage endogenous health care using a 12-month randomized controlled trial to demonstrate whether knee osteoarthritis (KOA)-at-risk individuals and KOA-diagnosed patients can improve their knee pain, physical function, and other relevant outcomes by using the technology to support knee/KOA self-management.

Knee osteoarthritis (KOA), the most common degenerative condition affecting the peripheral weight-bearing joints, leads to pain, restricted mobility, and financial and moral burden. These effects will increase with rising numbers of incidents, exacerbated by population aging and obesity. Zhang et al. found KOA in 43% and 22% (15% and 6% in symptomatic individuals) of Chinese women and men, respectively, 60 years or older. In the United States, KOA affects 12% of the elderly population, and it is projected that the need for total knee replacement (TKR) to treat end-stage KOA will grow by 673% to 3.48 million procedures a year from 2005 to 2030, with the demand for TKR revision doubling from 2005 to 2015. Hong Kong's public hospitals performed 6658 TKRs from 2011 to 2014, with 10,000 cases wait-listed for 2016. These figures suggest that KOA will impose significant and increasing pressure on Hong Kong's healthcare system.

As quadriceps muscle weakness increases the risk of KOA, its symptoms should be alleviated by leg muscle strengthening. Land-based exercise has been effective under close monitoring and supervision. A recent Cochrane Review involving 54 studies concluded that there is moderate to high-quality evidence that land-based exercises significantly improve muscle strength, physical function, and quality of life and reduce pain in KOA patients. The effect of such exercises is comparable to that of nonsteroidal anti-inflammatory drugs with no significant adverse effects. However, therapist-delivered exercise is costly and often impractical, especially in a public-health context, and Internet-delivered exercise regimes represent an alternative approach. In today's information technology (IT)-enabled environment, mobile apps are readily accessible to users of untethered devices (e.g., smartphones). The "any-where, any-time" mobile IT could be exploited to more effectively manage KOA by encouraging self-motivation to substitute for direct monitoring. This study will present "IT-centered endogenous healthcare" as a public policy to boost self-help in primary care and test its practicability and efficacy in KOA management in Hong Kong. The word "endogenous" suggests that self-motivation is a form of self-insurance in primary care. It is further suggested that demand-side incentives will be needed to induce individuals to incrementally allocate more resources (particularly time and effort) to incrementally self-insure health at the primary level.

On the supply side, IT-based healthcare products have been mainly designed for commercial considerations such as marketing and are insufficiently focused on function for purposes of public health and policy. We follow Liao-Cheung's approach to reify demand-side incentives in the form of a publicly funded (and hence free) mobile app, the adoption of which could encourage self-insurance among individuals with knee problems or KOA. The users' time and effort will initially be rewarded by the app's user-friendliness and lack of a fee. The study's technology intervention, "mobile rehabilitation (mRehab) app", will link smartphones to videos of evidence-based, physiotherapist-prescribed exercises to alleviate knee/KOA symptoms, together with educational (e.g., diet and behavioral modifications) and motivational components. This mobile IT platform will have easier and wider accessibility than exercises delivered via tethered devices (e.g., desktop computers), and it can engage the users' interest whenever and wherever KOA effects are felt. Over time, self-motivation is also enhanced by high-frequency prompts, periodic upgrades with feedback, support from Internet KOA-awareness groups, the possibility of accumulating one's own health-data to facilitate queries and dialogue, and (most importantly) monitoring the health progression predicted by the exercise regimen.

A natural research sample follows in the form of IT-enabled individuals with knee/KOA problems. The proposed hypothesis is then offered that sustained use of the mRehab app will significantly reduce the symptoms and improve functions of individuals with quadriceps weakness and knee pain or KOA over time and compared to a sham app (the analytical control). The effectiveness of the mRehab app regarding this hypothesis will be tested in two samples: (1) KOA-at-risk individuals and (2) KOA-diagnosed patients, in a randomized controlled trial (RCT).

Inclusion criteria for all participants:

• at least 50 years of age,
• a regular smartphone user,
• able to speak Cantonese and read Chinese,
• able to perform the exercises required in the physical assessments,
• able to provide written informed consent.

Inclusion criteria for KOA-at-risk participants:

• with subnormal quadriceps strength (as defined by the inability to complete five repetitions of a sit-to-stand test in less than 8 seconds [50 to 70 years of age], 10 seconds [70 to 79 years of age], or 12 seconds [80 years or older])
• with significant chronic knee pain (i.e., ≥2/10 on the 11-point Numerical Pain Rating Scale for more than 3 months on most days of the previous month, aggravated by weight-bearing or movement)
• without regular medical follow-up for knee problems

Inclusion criteria for KOA-diagnosed participants:

• with symptomatic radiographic KOA and being followed-up in hospitals
• have radiographic evidence of grade 2 to 3 KOA on the Kellgren-Lawrence scale in the posteroanterior and/or skyline view or the presence of lateral/posterior osteophytes.

Exclusion criteria:

• have a cognitive impairment,
• have nonambulatory status,
• have systemic inflammatory arthritis (e.g., gout),
• have a history of knee or hip replacement surgery,
• have a history of trauma or surgical arthroscopy of either knee within the past 6 months,
• involve in a similar study,
• participate in a knee exercise program within the past 6 months,
• have an intra-articular injection to the knee within the past 3 months,
• have recent or imminent surgery (within 3 months),
• have medical co-morbidities that preclude participation in exercise