**Background:** Overweight and obesity affect nearly two in three adults worldwide and are strongly associated with knee osteoarthritis (OA). Weight loss is recommended by OA clinical guidelines, yet few people with OA actively attempt to lose weight. Long-term weight maintenance after weight loss programmes is challenging, with research suggesting 60% of weight lost on VLCDs is regained over an average of 1.9 years. People with chronic musculoskeletal pain may experience more difficulty maintaining weight loss than those without pain. No previous studies had explored experiences of people with OA maintaining weight loss after a programme involving a ketogenic VLCD.
**Methods:** This qualitative study was nested within an RCT (ACTRN12618000930280) evaluating a weight loss programme for knee OA. Participants had completed a 6-month programme involving a ketogenic VLCD (approximately 800 kcal/day using Optifast or Optislim meal replacements), six videoconferencing consultations with a dietitian, six videoconferencing consultations with a physiotherapist, educational booklets, a Fitbit activity tracker, exercise bands, and a portion plate. Semistructured telephone interviews were conducted 6 months after programme completion (i.e., at 12 months from baseline) by a non-clinician researcher with no other involvement in the RCT. Interviews were audio recorded, transcribed verbatim, and analysed using reflexive thematic analysis following Braun and Clarke's six-phase approach. Twenty participants were interviewed (10 male, 10 female; mean age 65 years, SD 9.0; mean BMI 32.7 kg/m², SD 2.9; 60% from metropolitan areas). Participants had lost an average of 11.8% (SD 5.2) of body weight at programme completion and 9.7% (SD 6.9) at the time of interview.
**Key Results:** Three themes were developed: (1) successfully maintained weight loss — most participants described overall success, with some continuing to lose weight (1–4 kg), many maintaining weight, and some experiencing small gains (1–4 kg). Two participants (P5 and P13) regained significant weight (7–16 kg). Participants reported improvements in knee pain, self-confidence, and other health conditions like high blood pressure. (2) Empowering self-management of weight — five subthemes emerged: understanding importance of exercise and physical activity (participants engaged in walking, running, swimming, sports); increased knowledge about food and nutrition (from dietitian consultations and educational resources, leading to mindful eating and portion control); resources from programme still useful (booklets, recipe books, Optifast products, and recalled dietitian advice); knee pain as a motivator (motivation to avoid worsening pain or joint replacement surgery, though some with improved pain struggled with motivation); and confidence in ability to self-regulate weight (participants felt confident using VLCD principles again if needed). (3) Challenges keeping on track — three subthemes: loss of accountability to dietitian and study (drive and motivation waned without check-ins; participants desired follow-up consultations, phone calls, or emails); old habits and social situations (return to high-calorie foods, difficulty resisting carbohydrates, challenges with social eating); and stressful life events or changes in health (other health problems like hip replacement, back pain, appendicitis; COVID-19 pandemic leading to boredom snacking and reduced physical activity).
**Clinical Implications:** This is the first study to explore weight loss maintenance experiences specifically in people with OA. Findings suggest that a programme incorporating dietitian and physiotherapist consultations, a ketogenic VLCD, and educational/behaviour change resources supports confidence in maintaining weight loss in the medium term. Participants valued the combination of dietary and exercise components, supporting OA guideline recommendations for multimodal weight management. The loss of accountability to clinicians was a key challenge, suggesting that longer-term follow-up (booster sessions, digital health support, or peer group support) may be beneficial. Notably, some participants continued using meal replacement products and ketogenic diet principles long-term, which raises concerns about potential negative effects on electrolyte balance, hypotension, and gallstones, as well as disordered eating behaviours. Greater emphasis on transitioning to long-term healthy eating may be needed. The finding that improved knee pain reduced motivation for some participants highlights the need for ongoing goal redefinition and development of both extrinsic and intrinsic motivators. Study limitations include potential positive selection bias (participants volunteered for the RCT and qualitative study), the 6-month follow-up timeframe (longer-term outcomes unknown), and the sample being limited to English-speaking Australians with private health insurance, which may limit transferability to other settings.