Does Weight Loss Reduce the Severity and Incidence of Psoriasis or Psoriatic Arthritis?

A Critically Appraised Topic

S.K. Mahil; S.M. McSweeney; E. Kloczko; B. McGowan; J.N. Barker; C.H. Smith


The British Journal of Dermatology. 2019;181(5):946-953. 

In This Article


A systematic literature search was conducted using databases (MEDLINE, Embase and CENTRAL) and clinical trials registries. Search strategies combined free-text terms including 'psoriasis', 'psoriatic arthritis', 'obesity', 'weight loss' and corresponding medical subject headings to identify studies (Table S1; see Supporting Information). Secondary searching of bibliographies was also performed.

Full-text reports of randomized studies, nonrandomized studies (excluding case reports) and systematic reviews assessing interventions in obese (BMI > 30 kg m−2) or overweight (BMI > 25 kg m−2) populations published in English were included. In line with the National Institute for Health and Care Excellence (NICE) guidance on obesity management,[18] we considered interventions under three broad categories: lifestyle interventions (changes to diet or physical activity), pharmacological treatments [U.K.-licensed antiobesity medications including orlistat and glucagon-like peptide (GLP)-1 agonist antidiabetic medications that are associated with weight loss] and bariatric surgery.[19] Weight loss was categorized as mild (< 5%), moderate (5–10%) or significant (> 10%).[20] Mild and moderate-to-severe psoriasis were defined by baseline PASI < 10 and PASI > 10, respectively.[21]

Outcomes assessed were (i) incidence of psoriasis and/or PsA and (ii) change in psoriasis and/or PsA severity. To calculate incidence, only newly diagnosed or newly coded cases of psoriasis and/or PsA made in primary or secondary healthcare settings were considered. Acceptable measures of disease severity in psoriasis included mean change in PASI (ΔPASI), proportion of participants achieving PASI 75 (≥ 75% reduction in PASI from baseline) or change in Physician's Global Assessment.[22] Measurements used for disease severity in PsA were change in tender or swollen joint scores or achievement of minimal disease activity (MDA).[23]

Screening, data extraction and critical appraisal of search results (Figure 1) were performed independently by two authors (S.M.M. and E.K.).[24] Quality assessment and appraisal tools used were the Cochrane Risk of Bias Assessment Tool (randomized and nonrandomized studies with a prospective cohort design) and the Preferred Reporting of Case Series in Surgery (PROCESS) guidelines (case series).[25] Any disparities in risk-of-bias assessments were discussed and consensus reached.[26]

Figure 1.

PRISMA diagram outlining the search strategy.24 Screening of search results was performed by two authors (S.M.M. and E.K.). After duplicate studies were excluded, studies were screened by title and abstract. Full-text screening confirmed eligibility for inclusion in the final analysis. PsA, psoriatic arthritis; RCT, randomized control trial.

Meta-analyses of relevant studies were planned and conducted using Review Manager software (version 5·3; The Nordic Cochrane Centre, Copenhagen, Denmark). Random-effects models were used to account for high clinical and methodological variation between studies. When ΔPASI and its SD were not available by contacting the original authors, values were imputed from baseline and final values.[26] Sensitivity analyses were performed (Table S2; see Supporting Information) and investigations of heterogeneity were planned.