Study Design, Attributes & Descriptives Flashcards
(30 cards)
What overall study design did Zhuang et al. (2024) use?
A large-scale primary prospective cohort study using UK Biobank data.
Which UK resource supplied participants?
The UK Biobank, a population-based cohort of >500 000 adults aged 40–69 y.
What were the primary aims of the study?
- Examine associations between overall (PDI), healthful (hPDI) and unhealthful (uPDI) plant-based diet indices and incident CVD;
- explore potential mediators (dietary components & serum biomarkers).
(ii) explore potential mediators (dietary components & serum biomarkers).
Which three diet quality scores were analysed?
PDI, hPDI (healthful), and uPDI (unhealthful).
How were those indices constructed? (short answer)
Intake of 18 food groups was ranked into quintiles (1–5 pts); positive vs reverse scoring created PDI, hPDI, uPDI (score range 18–90).
Baseline data-collection window for dietary recalls?
2009-2012 (Oxford WebQ online 24-h recalls, up to 5 per person).
Final censor date for follow-up?
31 December 2020.
Mean follow-up duration (prediabetes vs diabetes)?
9.9 y (prediabetes) and 9.6 y (diabetes).
Initial ‘at-risk’ pool before exclusions?
~60 540 with prediabetes and 31 422 with diabetes identified in UK Biobank.
Give four major exclusion criteria at the 2nd screen.
Pre-existing CVD, cancer, missing/invalid 24-h recall, type 1 diabetes, or extreme energy intake (<600 / >4200 kcal men; <500 / >3600 kcal women).
Final analytic sample sizes?
17 926 with prediabetes; 7 798 with diabetes.
Attrition rates for each group?
Prediabetes ≈ 70.4 %; Diabetes ≈ 75.2 %.
Participant age range at baseline?
40–69 years.
Primary outcome definition?
Incident CVD (hospital ICD-10 codes for CHD, stroke, atrial fibrillation).
Key data-collection tool for diet?
Oxford WebQ self-administered 24-h recall.
How many food/drink items are in Oxford WebQ?
Nearly 200 distinct items.
Why were dietary recalls averaged across occasions?
To better reflect long-term intake and reduce measurement error.
Which index always assigns reverse scores to animal foods?
All three (PDI, hPDI, uPDI) assign reverse scores to animal foods.
What differentiates hPDI from uPDI scoring?
hPDI gives positive points to healthy plant foods & reverse to less-healthy; uPDI does the opposite.
Give two examples of ‘healthy plant foods’ in this study.
Whole grains; fruits/vegetables (nuts, legumes, tea/coffee also counted).
Give two examples of ‘less-healthy plant foods’.
Sugar-sweetened beverages; refined grains (plus potatoes, sweets/desserts, fruit juice).
What serum biomarkers were later tested as mediators?
Cystatin C, creatinine, CRP, HDL-C, IGF-1, among others.
Which risk assessment scores were reported for study quality?
Newcastle-Ottawa = 9/9; Downs & Black = 24/28; Jadad = 1/5.
Why exclude events within 2 years of baseline?
To reduce reverse causality (diet change due to pre-clinical CVD).
