Our study took place in the public hospitals in southern Adelaide, Australia, with a population of approximately 350,000. Between 28 September 2005 and 12 July 2006 (41 weeks), patients who had an initial consultation with an orthopaedic surgeon concerning a potential hip or knee replacement and who were then added to the waiting list for joint replacement surgery were invited to take part in the study.

Eligibility for inclusion was assessed on the day of the orthopaedic outpatient consultation by the project officer. Participants were eligible for inclusion if they: had a mini-mental score ≥24¹¹, were able to read and speak English, lived in the southern region of Adelaide, were not classified as requiring urgent surgery, and did not have significant frailty or illness that precluded completion of the protocol.

The research and ethics committees at Flinders Medical Centre and Repatriation General Hospital approved the study. The trial was performed in compliance with the Declaration of Helsinki.

Eligible patients were invited to participate and written informed consent was obtained from those who agreed to take part in the study. Once baseline assessments were complete, participants were randomly allocated to either receive the intervention or usual care (control), and this occurred within 24 h of giving consent. Allocations to the two treatment arms were computer-generated and randomised in blocks of 12. The allocation ratio was 1:1 (intervention:control).

A statistician external to the study generated the randomisation sequence using the random number generator in Microsoft Excel and created sequentially numbered, opaque, sealed envelopes containing group allocation for participants. The trial nurse enrolled the participants, and the clinical trials pharmacist who managed the randomisation allocation assigned participants to the groups. Block size remained unknown by trial staff until the cessation of recruitment.

The intervention was multi-faceted, and individuals could choose between zero and three of: the patient education self-management (the Partners in Health interview), a traditional education component (included at the request of the partner organisation Arthritis Foundation of South Australia Inc (Arthritis SA)), and peer support telephone calls. Within 7 days of random allocation to the intervention group, participants had one face-to-face session of goal setting with a research nurse (a senior registered nurse) using a structured approach that led to the formulation of a care plan. The structured approach used the “Partners in Health” model of self-management assessment and care planning to discuss participants' goals and strategies for managing their arthritis¹². For example, a participant with a goal to increase their walking capacity would make a detailed plan, that may address pain management and exercise along with community sources of relevant information. This protocol has been used in diabetes and respiratory patients previously¹². Monthly telephone calls to monitor participants' health status and reinforce the strategies for managing participants' arthritis were made using a standardized protocol by volunteer peer support educators from the local arthritis support and advocacy group Arthritis SA. The peer educators all had osteoarthritis, and had previously undertaken the disease specific modules, chronic disease modules and other peer educator training offered by Arthritis SA. In addition those participants in the intervention group could access two disease specific modules (one peer led, one clinical staff) from Arthritis SA for 2.5 h over 2 weeks.

Participants allocated to the control group received the usual standard of care for patients on orthopaedic waiting lists. Participants in the control group did not receive any additional educational material. Table I describes the components of the intervention and the usual care groups.

All individuals on an orthopaedic waiting list in South Australia have access to a generic chronic disease self-management course “Moving Towards Wellness” run by Arthritis SA. This course is based on Stanford University group based self-management programs¹³ and runs for 2.5 h weekly for 6 weeks. To monitor contamination, we assessed uptake of this course in both the intervention and control groups.

The information collected from participants included socio-demographic variables (age, gender, residential status, ethnic origin), quality of life, medication beliefs, a measure of pain and disability associated with osteoarthritis, depressive symptoms, and the impact of the educational strategies.

The Health Education Impact Questionnaire (HeiQ^{14, 15}) version 1 was designed to measure the impact of the self-management courses and was used as the primary outcome measure of the combined intervention components. The HeiQ is a reliable and validated 42 item questionnaire with eight subscales that reflect: positive and active engagement in life, health-directed behaviour, skill and technique acquisition, constructive attitudes and approaches, self-monitoring and insight, health service navigation, social integration and support, and emotional well-being. Each HeiQ subscale is scored between 1 and 6 with higher scores indicative of a more positive response. Construct validity for each subscale of the HeiQ has been previously demonstrated to be high through stringent confirmatory factor analysis and Cronbach's alpha for each scale has been found to be between 0.70 and 0.89¹⁵. Quality of life was measured using the Assessment of Quality of Life (AQoL)^{16, 17}. In the AQoL, a utility score is derived from scores on five dimensions that measure illness, independent living, social relationships, physical senses, and psychological well-being. The overall AQoL is scored on a life–death scale, where the lowest possible value, −0.04, represents quality of life states that are valued worse than death, 0.00 represents death-equivalent quality of life states and the highest value, 1.00, represents the best possible quality of life state¹⁶. The Beliefs about Medicines questionnaire (BMQ¹⁸) is a patient-centred measure that captures individual's general beliefs about use of prescription and non-prescription medicines. The Western Ontario and McMaster Universities (WOMAC) Arthritis Index¹⁹ was used to capture the pain (no pain 0; maximum pain 20), stiffness (no stiffness 0; maximum stiffness 8) and function (no disability 0; maximum disability 68) associated with osteoarthritis. Symptoms of depression were measured using the Centre for Epidemiologic Studies Depression Scale (CES-D), a 20-item scale with scores of ≥16/60 indicative of depressive symptomatology²⁰.

Six months after randomisation all participants were mailed a set of the same questionnaires that they completed at baseline and asked to complete and return them in the included reply paid envelope. We extracted information on whether joint replacement surgery had occurred within 6 months of randomisation from the hospital information systems. We enumerated the number of phone calls initiated by the peer support educators and the frequency of phone contact with participants. During calls, peer support educators utilized self-management action plans to identify progress against physical, emotional and social interventions. Goals were measured through a rating scale that looked at the level of success in progress towards goal achievement.

A total of 238 people were approached concerning participation in the study. Of these, 192 people met eligibility criteria and 152 provided consent and were subsequently randomised to the intervention (n=75) or control group (n=77). As shown in Fig. 1, there were a range of reasons for which 46 people were ineligible, including inability to read or speak English (n=16), rural location (n=12), requiring urgent surgery (n=11) or cognitive impairment (n=7).

Fig. 1

Recruitment flow of participants.

 

At baseline, there were no significant differences between the participants who subsequently received the intervention compared with those who were in the control group (Table II). The average age of participants was 67.5 years (SD 10.5) and 60 participants (39.5%) were male. Nearly half of the participants had completed 8 years or less of education. The median number of medical conditions (including osteoarthritis) was two. The average CES-D score at baseline was 15.0 (SD 9.8), and 54 participants (35%) had a CES-D score of 16 or more, suggestive of depressive symptomatology²⁰.

Sixty-six of the 75 participants (88%) allocated to the intervention completed a Partners in Health interview with a study nurse. A total of 39 members (52%) of the intervention group undertook the educational course at Arthritis SA, and of these 72% completed at least one additional module. Three members (4%) of the control group undertook an educational course, and none of these went on to complete an additional module. An average of 192.3 (SD 43.5) days elapsed between the date of randomisation and the date at which the follow-up questionnaires were returned.

At 6 month follow-up, the intention to treat analyses demonstrated that there was a significant difference in the primary endpoint of health-directed behaviour on the HeiQ between the participants who were in the intervention group (mean 4.29, 95% CI 3.99–4.58) and those in the control group (mean 3.81, 95% CI 3.52–4.09; P=0.017). In terms of secondary endpoints, there was also a significant effect of the intervention on skill and technique acquisition (mean 4.37, 95% CI 4.19–4.55) in comparison to control participants (mean 4.11, 95% CI 3.93–4.29; P=0.035; see Table III). Stiffness, as measured by the WOMAC, was significantly less at follow-up among participants in the intervention group (mean 6.6, 95% CI 6.1–7.0) vs those in the control group (mean 7.2, 95% CI 6.8–7.7; P=0.042). There was no significant effect of the intervention on the remaining HeiQ subscales, WOMAC pain or disability, AQoL, CES-D, or BMQ.

Twenty-two participants (11 in each group) had joint replacement surgery within the follow-up period, with an average of 88.4 days (SD 47.0) between randomisation and surgery for these participants. The follow-up questionnaires for these participants were completed an average of 106.3 days (SD 23.1) after surgery.

All 75 participants in the intervention group received at least one phone call but 25 people requested no further phone calls. Of those who received more than one call the average number of calls made to each participant was 5.2 (SD 3.3). During the study, 254 phone calls were initiated from the call centre, and 196 were successful.