# Effect of intensive blood pressure control on cognitive outcomes and brain structures in adults with type 2 diabetes: prespecified analysis of secondary outcomes of BPROAD randomised controlled trial.

- **DOI:** [10.1136/bmj-2026-101209](https://doi.org/10.1136/bmj-2026-101209)
- **Rating:** Strong (4.1 out of 5)
- **Verdict:** A large randomised trial finds no significant cognitive difference between blood pressure targets in type 2 diabetes, with long-term effects unresolved.
- **Framework version:** v5.13
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-11-19
- **Evaluation page:** https://www.nabu.science/paper/10.1136%2Fbmj-2026-101209

## Key claims

1. no statistically significant difference in incidence of a composite of all cause dementia or mild cognitive impairment was observed among adults with type 2 diabetes receiving standard treatment of SBP and intensive treatment
   - Support: Well-supported
   - Type: Primary finding
   - Basis: randomised comparison with masked adjudication and achieved event accrual supports the stated nonsignificant result over the observed follow-up
   - Location: Abstract Conclusions; Results, Table 2, Fig 3
2. intensive SBP treatment was associated with a reduction in the accrual of cerebral white matter lesion volume in the MRI substudy
   - Support: Tentative
   - Type: Secondary
   - Basis: selected MRI participation and 29.8% missing repeat scans leave the estimate dependent on weighting and imputation assumptions
   - Location: Conclusions
3. no statistically significant between group differences were observed for changes in global cognitive function (mean between group difference 0.011 standardised units, 95% CI -0.009 to 0.031; P=0.28)
   - Support: Well-supported
   - Type: Secondary
   - Basis: longitudinal randomised comparison with high assessment retention and consistent sensitivity analyses supports the reported nonsignificant difference
   - Location: Results, Cognitive assessment; Table 3
4. No statistically significant between group difference was observed for changes in total brain volume (between group difference 2.97 cm³, 95% CI -0.86 to 6.81; P=0.13) (table 4).
   - Support: Supported
   - Type: Secondary
   - Basis: selected substudy participation and missing repeat scans limit the scope of the reported nonsignificant comparison
   - Location: Results, MRI outcomes; Table 4

## Strengths

- **Large trial addresses a diabetes evidence gap** The cognitive cohort included 12,434 adults and accrued 1,530 adjudicated composite events, exceeding the 1,374-event target. It supplies a direct comparison in a population excluded from SPRINT MIND and assessed less comprehensively in ACCORD MIND.
  - Dimension: contribution
  - Location: Introduction; Statistical analysis; Table 2
- **Masked adjudication supports the randomised comparison** Central randomisation and masked cognitive adjudication support outcome assessment despite open-label treatment. Multiple-imputation, competing-risk, and alternative missing-data analyses examine important sources of uncertainty.
  - Dimension: methodological rigour
  - Location: Randomisation and masking; Adjudication and cognitive outcomes; Statistical analysis
- **Discussion directly engages divergent trial findings** The Discussion examines differences from SPRINT MIND and ACCORD MIND rather than treating the results in isolation. It considers baseline cognitive status, follow-up duration, and possible medication-related explanations for differing brain-volume findings.
  - Dimension: positioning
  - Location: Discussion, Comparison with other studies

## Limitations

- **MRI inference depends on incomplete selected follow-up** Only 605 of 862 participants completed repeat MRI, leaving selection and imputation assumptions important to the positive finding. Table 4 labels follow-up populations as 419 and 428 rather than 302 and 303 observed scans, and its printed asinh formula is incorrect, limiting reproducibility without proving an analysis error.
  - Dimension: methodological rigour
  - Location: Results, MRI outcomes; Figure 5; Table 4 and footnote
- **Adverse-event results are not reported here** This analysis describes safety monitoring but does not present systematic adverse-event results. Its cognitive findings therefore cannot independently establish the overall benefit-harm balance of intensive treatment.
  - Dimension: methodological rigour
  - Location: Methods, Study design and participants; Results; Introduction, reference to the primary cardiovascular report
- **Cognitive inference remains bounded by ascertainment** Mild cognitive impairment was operationalised through persistent decline from each participant's baseline, introducing uncertainty about endpoint sensitivity. A median 3.8 years of follow-up also leaves delayed cognitive effects unresolved.
  - Dimension: methodological rigour
  - Location: Adjudication and cognitive outcomes; Discussion, Comparison with other studies and Research implications

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted
  - "Follow-up(95% CI); n=419" (Table 4, Follow-up column header) vs "605 (70.2%) participants completed a repeat MRI scan (302 intensive treatment v 303 standard treatment)" (Results, MRI outcomes). Table 4 labels follow-up populations as 419/428 rather than the 302/303 observed repeat scans, without explicitly identifying the headers as imputed analysis populations. The effect on the result is unclear.
- **Reporting and transparency:** Noted. The printed inverse hyperbolic sine formula is incorrect, leaving the implemented transformation uncertain from the description alone. Ethics approval, consent, registration, competing interests, and a named data and code repository are declared.
  - Conflict of interest: declared · Ethics approval: declared · Preregistration: declared · Data availability: declared · Code availability: declared
- **Reference integrity:** Noted
  - 43 references in manuscript
  - 39 of 39 checkable references found in an index
  - 4 could not be checked — not the same as no match found
  - 4 mistyped DOIs matched to the cited work by its details
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-11-19
