General Medicine · Medical Literature
Remote person-centred care and long-term medication management in primary care: post hoc analysis of a randomised controlled trial.
Abstract
To explore refill adherence to prescribed treatment for chronic heart failure (CHF) and/or chronic obstructive pulmonary disease (COPD), during and after a remote person-centred care (PCC) intervention. Post hoc analysis of a randomised controlled trial cohort. We used data from the PROTECT (Person-Centred Care at Distance) randomised controlled trial in Swedish primary care, which were linked to data from the Swedish Prescribed Drug Register. A total of 222 participants from nine primary care centres were enrolled in PROTECT. This study included participants with dispensed medications for CHF and/or COPD, excluding those with multidose dispensing (intervention n=50; control n=49, 42% women, mean age 69 years). PROTECT evaluated a PCC intervention using a digital platform and structured telephone support in addition to usual care versus usual care alone. Refill adherence was compared between groups, defining appropriate medication supply as ≥80% of days covered over 2 years. We also examined associations between refill adherence and number of long-term conditions and medications. Refill adherence did not differ significantly between groups. A higher number of long-term conditions was associated with lower odds of refill adherence (adjusted OR (aOR) per additional condition: 0.71; 95% CI 0.50 to 0.97; p=0.041), as were more long-term medications (aOR per additional medication class: 0.84; 95% CI 0.71 to 0.98; p=0.029). The PROTECT intervention did not improve refill adherence, suggesting that future person-centred interventions may require more structured medication-management focus. Increasing numbers of long-term conditions and medications were both associated with lower refill adherence. NCT03183817.
Clinical relevance
Does a remote person-centred care (PCC) intervention using a digital platform and structured telephone support improve refill adherence to medications for chronic heart failure (CHF) and/or chronic obstructive pulmonary disease (COPD) in primary care, and what patient-level factors are associated with refill adherence? __SNAPSHOT__ {"studyDesign":"Post hoc analysis of a randomised controlled trial cohort (parent trial: PROTECT — Person-Centred Care at Distance)","setting":"Nine primary care centres in Sweden (Swedish Prescribed Drug Register linked to trial data)","sampleSize":"n=99 (intervention: n=50; control: n=49); drawn from the parent PROTECT trial of 222 participants","population":"Adults with dispensed medications for CHF and/or COPD enrolled in the PROTECT RCT; 42% women, mean age 69 years; patients on multidose dispensing systems were excluded","intervention":"Remote person-centred care (PCC) comprising a digital platform and structured telephone support in addition to usual care","comparator":"Usual care alone","primaryOutcome":"Refill adherence, defined as achieving ≥80% of days covered (PDC) for CHF and/or COPD medications over a 2-year period; also examined associations between refill adherence and number of long-term conditions and long-term medications","followUp":"2 years (based on PDC calculation window)","fundingSource":"Not reported in the abstract; affiliated with the University of Gothenburg Centre for Person-Centred Care (GPCC) — presumed academic/government; confirm in full text"}
Key findings
Refill adherence did not differ significantly between the intervention and control groups (specific proportions and between-group p-value not reported in the abstract). In adjusted multivariable logistic regression (adjusted for age, sex, smoking status, education, and civil status), each additional long-term condition was independently associated with lower odds of refill adherence (aOR 0.71; 95% CI 0.50–0.97; p=0.041). Similarly, each additional long-term medication class (ATC level) was associated with lower odds of adherence (aOR 0.84; 95% CI 0.71–0.98; p=0.029). Descriptively, adherent patients had a lower median number of long-term conditions (median [IQR]: 3 [2–4]) versus non-adherent patients (4 [3–5]; Wilcoxon rank-sum p=0.023), and fewer long-term medication classes (adherent median [IQR]: 5 [3–7] vs. non-adherent: 7 [5–8]; Wilcoxon rank-sum p=0.006). NNT and NNH are not calculable from available data. Adverse events were not reported.
Practice implications
This post hoc analysis of the PROTECT RCT found no statistically significant effect of a remote PCC intervention on refill adherence to CHF and/or COPD medications over 2 years. The most clinically important finding is the inverse association between treatment complexity — both multimorbidity and polypharmacy — and refill adherence: patients with more conditions and more medication classes were less likely to achieve the ≥80% days-covered threshold. These associations, while statistically significant, are modest in absolute effect size (aOR ~0.71–0.84 per additional condition or drug class) and their clinical magnitude depends on baseline adherence rates not reported in the abstract. The findings apply to community-dwelling older adults (mean age 69) with CHF and/or COPD in Swedish primary care; patients requiring multidose dispensing systems were excluded, limiting generalizability to the most adherence-vulnerable patients. Because this is a post hoc, hypothesis-generating analysis — not a pre-specified adherence endpoint — causal inference is not warranted. Only the abstract was available for review; full-text data on adherence rates by group, subgroup analyses, and sensitivity analyses were not accessible. __LIMITATIONS__ Post hoc analysis: adherence was not a pre-specified primary endpoint of the PROTECT RCT, introducing risk of data dredging and inflated type I error.,Small analytic sample (n=99), substantially underpowered to detect between-group differences in refill adherence; the null result may reflect a type II error.,Exclusion of patients on multidose dispensing systems selectively removes the highest-risk, most adherence-challenged patients, limiting generalizability.,Refill adherence (pharmacy dispensing records) is a proxy measure and does not confirm actual medication ingestion.,The 2-year observation window spans both the active intervention phase and a post-intervention period; these are not reported separately, obscuring any temporal effect of the intervention.,Adjusted covariates (age, sex, smoking, education, civil status) may not fully account for confounding by indication, disease severity, or health literacy.,Single-country (Sweden) primary care setting with universal healthcare coverage limits external validity to other healthcare systems.,Funding source and potential conflicts of interest are not confirmed from the abstract alone. __BOTTOM_LINE_CATEGORY__ promising_but_preliminary __BOTTOM_LINE__ A remote person-centred care intervention delivered via digital platform and telephone support did not significantly improve refill adherence to CHF and/or COPD medications compared with usual care over 2 years in this small post hoc analysis. However, the study provides hypothesis-generating evidence that greater multimorbidity and polypharmacy are independently associated with lower refill adherence, identifying a clinically actionable target population. These findings should not change prescribing practice but should prompt prospective, adequately powered trials testing adherence-focused PCC strategies specifically in patients with high disease and medication burden.