Table 2

Causality and preventability of hospital admission associated with medication non-adherence

Author (year)Causality
Causality judgementCausal relationshipPreventability criteriaPreventability judgement% Preventability
McKenney et al (1976)19 NRNRNRNRNRNR
Stewart et al (1980)20 NRNRNRNRNRNR
Bergman and Wiholm (1981)21 Algorithm of Bergman and Wiholm21 By researchersDefinite or probable: (21/21; 100%)NRNRNR
Yosselson-Superstine and Weiss (1982)22 NRNRNRNRNRNR
Bigby et al (1987)23 No explicit criteriaBy three reviewers and consensus (at least two ‘yes’ judgements)Definite (26/26; 100%)No explicit criteriaBy three reviewers and consensus (at least 2 ‘yes’ judgements)19/26 (73.1)
Davidsen et al (1988)24 WHO criteria45 By research physicianDefinite, probable or possible (16/16; 100%)NRNRNR
Grymonpre et al (1988)25 Algorithm of Bergman and Wiholm21 By attending physicians and house staffNR for non-adherenceNRNRNR
Col et al (1990)2 Col2 By two senior medical residentsDefinite or probable (11/34; 32.35%), possible (13/34; 38.24%), contributing factor (10/34; 29.41%)NRNRNR
Stanton et al (1994)26 Hallas44 By an attending medical officer and a panel of four of the authorsDefinite (6/10; 60%), probable (4/10; 40%)NRNRNR
Courtman and Stallings (1995)27 NRNRNRCourtman and Stallings27 By pharmacy residents9/9 (100)
Dartnell et al (1996)28 Karch-Lasagna47 By at least two authors and discrepancy was resolved by all authorsNRDartnell28 By at least two authors,
discrepancy resolved by all
15/15 (100)
Nelson and Talbert (1996)29 Modified Hallas44 By investigatorsDefinite or probable
(48/48; 100%)
Nelson and Talbert29 By investigatorsNR
Murad and Chawla (1997)30 NRNRNRNRNRNR
Chan et al (2001)31 Hallas44 By investigators and training doctorsNRHallas preventability44 By investigators and trainee doctors9/9 (100)
Malhotra et al (2001)32 Malhotra32 By one of investigatorsNRNRNRNR
Martin et al (2002)33 Hallas44 By investigatorsDefinite, probable or possible (91/91; 100%)Schumock and Thornton49 By investigators91/91 (100)
Otero López et al (2006)34 Modified algorithm of Karch-Lasagna47 By investigatorsDefinite or probable (19/19; 100%)Schumock and Thornton49 By investigatorsNR
Samoy et al (2006)35 Using explicit predefined approachBy three reviewers and consensusReport only DRP resultZed50 and Forster51 By three reviewers and consensus22/22 (100)
Kongkaew (2009)36 Hallas,44 amended Howard48 By three reviewers and consensusDefinite or probable (190/190; 100%)Hepler and Strand43 By three reviewers and consensus56/190 (29.47)
Singh et al (2011)37 NRNRNRZed50 and Forster51 By investigators55/55 (100)
Al-Arifi et al (2014)38 NRNRNRNelson and Talbert29 By investigators17/17 (100)
Kongkaew (2015)39 Hallas,44 amended Howard48 By three reviewers and consensusCausal: overuse (11/32; 34%), underuse (21/32; 66%)Hepler and Strand43 By three reviewers and consensus32/32 (100)
Gustafsson et al (2016)40 Using WHO criteria46 Using explicit criteriaDefinite (7/19; 36.8%), probable (5/19; 26.3%), possible (7/19; 36.8%)NRNRNR
Jolivot et al (2016)41 NRNRNRSchumock and Thornton49 Judged by investigators31/31 (100)
  • DRP, drug-related problem; NR, not reported.