TY - GEN
T1 - Prognostic factors in Chronic Inflammatory Diseases
T2 - A PhD study within clinical epidemiology
AU - Overgaard, Silja Hvid
PY - 2023/9/13
Y1 - 2023/9/13
N2 - Background: Chronic inflammatory diseases (CIDs) associated with the Western lifestyle include
Crohn’s disease, ulcerative colitis, rheumatoid arthritis, axial spondyloarthritis, psoriatic arthritis and
psoriasis and are all closely related. These diseases are a considerable burden for patients and society;
unfortunately, there is no cure. Thus, the therapeutic goal is to relieve symptoms and prevent disease
progression and morbidity. Biological disease-modifying drugs have revolutionised the treatment of
CIDs, but remission rates are less than optimal.Aim: Therefore, the generic aim of this PhD was to improve the treatment of these CIDs by
identifying dietary prognostic factors and ensuring trustworthy evidence. The aims of the individual
studies were: I. To investigate if the treatment response of biological therapy is associated with habitual
dietary intakes of fibre and red/processed meat in patients with CIDs
II. To identify and characterise dietary patterns among patients with CIDs and investigate
whether patient characteristics were associated with these patterns
III. To explore the impact of attrition rates in randomised trials of patients with CIDs treated
with biological and targeted synthetic disease-modifying drugsMethods: We prospectively enrolled patients with one of the six CID diagnoses and a plan to initiate
biological therapy. The patients were clinically assessed before and after 14-16 weeks of treatment.
At baseline, patients completed a food frequency questionnaire enabling estimation of habitual dietary
intake. Based on the cohort’s longitudinal data, we stratified patients into two groups determined by the ratio
of fibre to red/processed meat (a high fibre/low meat group and a low fibre/high meat group) to
compare the proportion of clinical response after 14-16 of treatment (study I).In a cross-sectional analysis of the baseline data, we identified and characterised dietary patterns and
the dietary pattern membership for each patient using explorative factor, latent profile and descriptive
analyses (study II).In a meta-epidemiological study, we synthesised randomised trials testing the effect of biological and
synthetic targeted disease-modifying drugs in patients with CIDs (study III). We extracted
information on treatment effects and attrition and combined the trials using meta-regression models
to analyse the associations between treatment effects and attrition.ResultsIn study I, we found that a high intake of ‘dietary fibre and low intake of red/processed meat’ was not
associated with better treatment response across the group of CIDs (OR=1.48, 95% CI 0.72 to 3.05).
However, the results indicated a beneficial effect among rheumatoid arthritis patients (OR=9.84, 1.35
to 71.56). In study II, we identified and characterised three dietary patterns, i.e., the ‘Meat and Salt’, the
‘Vegetables’ and the ‘Mixed’ dietary pattern, which varied in nutrient profiles but were similar in patient characteristics. Common for all three patterns were excessive and deficient intake of nutrients
typical of the Western diet.In study III, differential attrition prevailed among randomised trials, with more attrition in control
than intervention groups (OR=0.45, 0.41 to 0.50). The odds of treatment benefit were 4.39 (3.94 to
4.89) higher in the intervention groups. This estimate was inversely associated with the OR for
differential attrition, meaning that trials with the most attrition in the control groups were connected
to better treatment effect estimates.Conclusion: In conclusion, CID patients do not generally eat according to recommendations for
healthy eating and prevailing dietary patterns among them were not associated with specific
phenotypes. A high intake of fibre in combination with a low intake of red/processed meat did not
affect the clinical response to biological treatment across the six CIDs but showed a beneficial effect
among RA patients, warranting confirmation in randomised trials. Targeting diet as one of the modifiable risk factors for treating CIDs has promising potentials, but
evidence-based guidelines are urgently needed. The validity of research findings must be critically
appraised to establish such an evidence base, which encompasses assessing the risk of attrition bias,
including evaluating the presence of differential attrition.
AB - Background: Chronic inflammatory diseases (CIDs) associated with the Western lifestyle include
Crohn’s disease, ulcerative colitis, rheumatoid arthritis, axial spondyloarthritis, psoriatic arthritis and
psoriasis and are all closely related. These diseases are a considerable burden for patients and society;
unfortunately, there is no cure. Thus, the therapeutic goal is to relieve symptoms and prevent disease
progression and morbidity. Biological disease-modifying drugs have revolutionised the treatment of
CIDs, but remission rates are less than optimal.Aim: Therefore, the generic aim of this PhD was to improve the treatment of these CIDs by
identifying dietary prognostic factors and ensuring trustworthy evidence. The aims of the individual
studies were: I. To investigate if the treatment response of biological therapy is associated with habitual
dietary intakes of fibre and red/processed meat in patients with CIDs
II. To identify and characterise dietary patterns among patients with CIDs and investigate
whether patient characteristics were associated with these patterns
III. To explore the impact of attrition rates in randomised trials of patients with CIDs treated
with biological and targeted synthetic disease-modifying drugsMethods: We prospectively enrolled patients with one of the six CID diagnoses and a plan to initiate
biological therapy. The patients were clinically assessed before and after 14-16 weeks of treatment.
At baseline, patients completed a food frequency questionnaire enabling estimation of habitual dietary
intake. Based on the cohort’s longitudinal data, we stratified patients into two groups determined by the ratio
of fibre to red/processed meat (a high fibre/low meat group and a low fibre/high meat group) to
compare the proportion of clinical response after 14-16 of treatment (study I).In a cross-sectional analysis of the baseline data, we identified and characterised dietary patterns and
the dietary pattern membership for each patient using explorative factor, latent profile and descriptive
analyses (study II).In a meta-epidemiological study, we synthesised randomised trials testing the effect of biological and
synthetic targeted disease-modifying drugs in patients with CIDs (study III). We extracted
information on treatment effects and attrition and combined the trials using meta-regression models
to analyse the associations between treatment effects and attrition.ResultsIn study I, we found that a high intake of ‘dietary fibre and low intake of red/processed meat’ was not
associated with better treatment response across the group of CIDs (OR=1.48, 95% CI 0.72 to 3.05).
However, the results indicated a beneficial effect among rheumatoid arthritis patients (OR=9.84, 1.35
to 71.56). In study II, we identified and characterised three dietary patterns, i.e., the ‘Meat and Salt’, the
‘Vegetables’ and the ‘Mixed’ dietary pattern, which varied in nutrient profiles but were similar in patient characteristics. Common for all three patterns were excessive and deficient intake of nutrients
typical of the Western diet.In study III, differential attrition prevailed among randomised trials, with more attrition in control
than intervention groups (OR=0.45, 0.41 to 0.50). The odds of treatment benefit were 4.39 (3.94 to
4.89) higher in the intervention groups. This estimate was inversely associated with the OR for
differential attrition, meaning that trials with the most attrition in the control groups were connected
to better treatment effect estimates.Conclusion: In conclusion, CID patients do not generally eat according to recommendations for
healthy eating and prevailing dietary patterns among them were not associated with specific
phenotypes. A high intake of fibre in combination with a low intake of red/processed meat did not
affect the clinical response to biological treatment across the six CIDs but showed a beneficial effect
among RA patients, warranting confirmation in randomised trials. Targeting diet as one of the modifiable risk factors for treating CIDs has promising potentials, but
evidence-based guidelines are urgently needed. The validity of research findings must be critically
appraised to establish such an evidence base, which encompasses assessing the risk of attrition bias,
including evaluating the presence of differential attrition.
U2 - 10.21996/q7bb-py18
DO - 10.21996/q7bb-py18
M3 - Ph.D. thesis
PB - Syddansk Universitet. Det Sundhedsvidenskabelige Fakultet
ER -