An expert-supported monitoring system for patients with chronic obstructive pulmonary disease in general practice: results of a cluster randomised controlled trial

• patient diagnosed with COPD or asthma with persistent airflow obstruction, as confirmed with the patient’s most recent spirometry (forced expiratory volume in 1 second [FEV1]/forced vital capacity [FVC] < 70%, or postbronchodilator FEV1 < 80% predicted and ≥ 9% reversibility);11,12

• the patient’s lung function data from the previous year were available at the general practice diagnostic centre; and

• patient aged at least 25 years.

Exclusion criteria were:

• patient treated by a chest physician;

• patient participating in another respiratory intervention study;

• GP considered it detrimental to the patient to participate in the study;

• patient had any serious other non-pulmonary diseases (or disease stages) or pulmonary diseases (eg, sarcoidosis, lung cancer, lung fibrosis); or

• patient could not read.

GPs at practices who had referred more than six patients to the RDC were contacted and asked to participate. We used computerised minimisation to allocate practices to the monitoring and usual care groups13,14 while stratifying for:

• group versus solo practice;

• practice nurse employed versus no practice nurse employed; and

• ≤ 10 versus > 10 patients fulfilling the study inclusion criteria.

Step B. Information from the monitoring visit and previous visits was sent to a respiratory expert (chest physician or GP with special respiratory interest). The respiratory experts gave recommendations regarding treatment, additional diagnostic tests and referrals to other disciplines, based on national clinical practice guidelines for COPD and asthma.11,12 Experts’ interpretation based on spirometry results and written information has been shown to be valid.16

Step C. Written feedback was sent to the patient’s GP. The patient was instructed to visit the GP 2 weeks after the monitoring visit to discuss the outcome. During these visits, the expert recommendations could be implemented by the GP (eg, checking inhalation technique) or recommended to the patient. Half-yearly visits from a nurse consultant to the practice to support GPs in implementing the recommendations were an integral part of the expert-supported monitoring system.

Step D. Ultimately, the patient should implement the recommendations made (eg, quit smoking, increase exercise).

Participants completed questionnaires at baseline, at 1 year, and at the end of the study. The primary study outcome was the CRQ score.17 Secondary outcomes were: CRQ domain scores; generic health-related quality of life (physical and mental domains of the Short-Form 12 [SF-12] and the EuroQol-5D);18-20 breathlessness according to level of exertion (Modified Medical Research Council [MMRC] score,2 dichotomised as 0–1 and 2–4); occurrence rate of self-reported exacerbations; and annual FEV1 decline.

For the process evaluation, the respiratory experts’ database was examined to collect data on their recommendations. The nurse consultant collected data on GPs’ implementation of recommendations. Patient questionnaires comprised questions about disease management. At the end of the study, the nurse consultant collected information on disease management from GPs in the usual care group.

Multilevel repeated measurement regression analysis was used to model the effect of monitoring on CRQ overall score, CRQ domain scores, SF-12 scores, EuroQol-5D scores, FEV1 decline, and dichotomised MMRC scores. We used a PROC MIXED procedure in SAS statistical software (SAS Institute, Cary, NC, USA), with general practice as the random coefficient and compound symmetry correlation structure. Multilevel logistic regression analysis was used to analyse effects on exacerbations. All models were corrected for sex, age, socioeconomic status, baseline cigarette smoking status, reversibility, exacerbations at baseline, use of inhaled corticosteroids, use of long-acting bronchodilators, and postbronchodilator FEV1 % of the predicted value. Participants were included in the analysis if they participated in the study (intention-to-treat analysis).

Box 1 shows the process of practice and patient recruitment and follow-up. Thirty-four general practices participated. From these, 261 of 286 eligible patients (91%) responded to the invitation, and 213 (74%) were willing to participate. No significant differences between participants and non-participants with regard to sociodemographic characteristics, medication use, and spirometric indices were found. Twenty-four patients did not enter the study, and 19 patients were excluded from analyses. Data from 170 participants were used for the analyses. Box 2 shows the baseline characteristics of both groups.

The study was originally designed to evaluate monitoring of patients with COPD and asthma with a chronic airflow obstruction. However, after the recruitment phase we found that almost all of the patients fulfilled the criteria for COPD (ie, FEV1/FVC < 70% postbronchodilator); therefore, we decided to focus on COPD.

Box 3 shows the mean overall CRQ scores in the monitoring and usual care groups. Based on repeated measurement analyses, the additional gain in CRQ score during follow-up was 0.004 points for monitoring compared with usual care (95% CI, −0.172 to 0.180). Box 4 summarises the results for the secondary outcomes. No significant differences between the monitoring and usual care groups were observed other than CRQ domain mastery.

In total, respiratory experts gave 290 recommendations (Box 5). Smoking cessation was the most frequent recommendation (28% of all recommendations), and inhaler technique training and assessment of compliance with medical treatment were also recommended regularly. In 146 monitoring visits (50%), the respiratory experts did not consider any modification in disease management necessary. For 73 patients (89%), the GPs received at least one recommendation to change disease management.

Information about 274 of the 290 recommendations could be collected (Box 5). According to GPs, they attempted to implement 125 (46%) of the 274 recommendations. In practices with a practice nurse, the implementation rate of recommendations was higher than in practices without a practice nurse (P < 0.05); in particular, inhaler technique training and checking medication compliance were implemented more frequently (Box 5). The main reason why recommendations were not implemented was because patients did not visit the practice after their monitoring visit at the RDC. On 45 occasions, patients did not visit the GP after a monitoring visit that had resulted in at least one recommendation.

Changes in disease management are shown in Box 6. In general, only minor changes in disease management were achieved in the monitoring group, and these changes were comparable to usual care.