Understanding Artificial Pancreas Systems: Results from a 24-Month Trial

Related and Discussions

Subscribe now for the latest breakthroughs, expert insights, and cutting-edge updates in diabetes care—delivered straight to your inbox.

Stay informed. Stay ahead.

{"intervention_arm_id":"arm_1","reference_arm_ids":["arm_2"],"comparison_type":"between_group"}

HbA1c reduction was greater in intervention group (−2.57%) than control (−1.96%).

HbA1c fell more in the intervention group than in usual care over 12 months (−2.57% vs −1.96%; adjusted β −0.632, 95% CI −1.027 to −0.238; P<.001).

10-year ASCVD risk decreased significantly in the intervention group compared with control.

Estimated 10-year ASCVD risk decreased more with the collaborative care model than usual care (adjusted β −1.206, 95% CI −1.730 to −0.682; P<.001).

Fasting glucose values were lower at 12 months in the intervention group, but the adjusted group-by-time interaction was not significant (β −0.602; P=.27).

Two-hour postprandial glucose improved more in the intervention group than usual care (adjusted β −1.210 mmol/L; P=.03).

Waist circumference decreased more in the intervention group than usual care (adjusted β −3.557 cm; P<.001).

Waist-to-hip ratio decreased more in the intervention group than usual care (adjusted β −0.020; P<.001), supporting an effect on central adiposity.

Systolic blood pressure improved more with the intervention than usual care (adjusted β −3.556 mm Hg; P=.008).

Diastolic blood pressure improved more with the intervention than usual care (adjusted β −2.639 mm Hg; P=.01).

Triglycerides improved more with the intervention than usual care (adjusted β −0.814 mmol/L; P=.01).

Total cholesterol improved more in the intervention group than usual care (adjusted β −0.540 mmol/L; P=.001).

Satisfaction was 32.43 in intervention and 26.16 in control.

Treatment satisfaction was higher in the intervention group at 12 months (32.43 vs 26.16; adjusted β −4.314; P<.001). The beta direction reflects model coding, while raw scores favor the intervention.

Total cost was 5505.86 CNY in intervention and 7902.61 CNY in control; P=.01.

Total diabetes-related costs were lower in the intervention group than usual care over 12 months (5505.86 vs 7902.61 CNY; P=.01), mainly with lower hospitalization costs despite higher outpatient-related costs.

A multifaceted primary care model combining routine physician-led therapy with pharmacist care, in-person diabetes management visits, and biweekly WeChat education over 12 months.

Positive: 11; Negative: 0; No effect: 1; Mixed: 0

Direction counts recalculated from intervention outcomes due to mismatch with AI-provided counts.

Extracted 12 outcome signals across the study

Primary outcomes and several cardiometabolic, anthropometric, satisfaction, and cost outcomes favored intervention; fasting glucose, BMI, HDL, and LDL did not show clear between-group effects.

Direction counts were recalculated from extracted outcomes because the AI-provided totals did not match.

↓ HbA1c improved more with digital physician-pharmacist care than usual care (−2.57% vs −1.96%)

Mostly intermediate or surrogate outcomes

A structured pharmacist-physician model can improve HbA1c in primary care adults with type 2 diabetes

Remote education and in-person pharmacist visits may support broader cardiometabolic improvements

The model may reduce total diabetes-related costs, mainly through lower hospitalization costs

Xiao J, Wang Q, Tan S, et al. Effect of a Digital-Driven Physician-Pharmacist Collaborative Model for Diabetes in Primary Health Care: Cluster Randomized Trial. J Med Internet Res. 2026;28:e77470. doi:10.2196/77470

Participants and clinicians could not be blinded to the intervention

Only six primary care centers were randomized, which limits cluster-level precision

COVID-19 may have affected care-seeking behavior during the trial

Some secondary outcomes may have been underpowered

A 12-month cluster trial in 574 adults with type 2 diabetes found that digital pharmacist-physician care improved HbA1c, cardiovascular risk, waist measures, and satisfaction compared with usual physician care.

↓ 10-year ASCVD risk decreased compared with usual care (adjusted β −1.206)

↑ Treatment satisfaction was higher in the intervention group after 12 months

Routine physician-led therapy combined with pharmacist-delivered pharmaceutical care at quarterly visits and biweekly remote education via WeChat.

Digital-driven physician-pharmacist collaborative model

Patients in control centers received usual treatment and management by physicians.

Usual physician-led care

Includes self-reported outcomes (moderate): Self-reported outcomes used

Limited safety reporting (moderate): Safety reporting depth: limited

No post-intervention follow-up (moderate): 0 weeks of follow-up after treatment ended

Relies on surrogate outcomes (moderate): No hard clinical outcomes reported

Digital pharmacist-physician care improves HbA1c in type 2 diabetes

This study tested a digital physician-pharmacist care model in primary care centers in China. Over 12 months, the model improved HbA1c and cardiovascular risk more than usual physician care, while also improving several metabolic measures and patient satisfaction.

This was a cluster trial, so results depend partly on clinic-level implementation

The intervention used trained pharmacists and WeChat, which may not transfer directly to all settings

The study tested a care model, not a single medication or device

Several outcomes improved, but fasting glucose, BMI, HDL cholesterol, and LDL cholesterol did not show clear between-group effects

Digital pharmacist-physician care improves HbA1c in type 2 diabetes

Study at a glance

Main effects

Evidence Suggest

Who this applies to

Keep in Mind

Between the Lines

Unlock Full Analysis

Journal Reference

No ads. No tracking.

Secure & private

Always up to date