Characteristics of 24 included studies
| Author (year) | Country | Ward admitted, study setting | Study period | Population | Mean age, % male | Definition of medication (non)-adherence | Method of detection | HA rate due to NA (%) | Type of non-adherence (%) | HA rate due to NA among patients with MRP | Medication classes involved (%) | Quality assessment |
| McKenney and Harrison (1976)19 | USA | General medical ward, teaching hospital | 2 months | General population | 56.1% of population aged 50 years or over | Patients admit to taking less than the prescribed number of doses of a given medication during the month prior to hospital admission. | Interview and medical record review | 23/216 (10.65) | NR | 23/59 (38.98) | CVS drugs (39.1), endocrine drugs (26.1), CNS drugs (21.7), respiratory drugs (8.7), infections (4.3) | Moderate risk of bias |
| Stewart et al (1980)20 | USA | Inpatient psychiatric unit | 6 months | General population | 33.9 years, 40% male | Hospital admission was related to patients not following directions for prescribed medications. | Interview, medical record review and drug level analysis | 5/60 (8.33) | NR | 5/25 (20.00) | CNS drugs (80), endocrine drugs (20) | Moderate risk of bias |
| Bergman and Wiholm (1981)21 | Sweden | Medical ward, university hospital | 3.5 months | General population | 59±19 years (range 16–97 years), 49.5% male | Haynes definition42 as the extent to which the patient behaviour coincided with clinical prescription. | Interview, medical record review and drug level analysis | 21/285 (7.37) | UD (11/21), OD (10/21) | 21/45 (46.67) | NR | Low risk of bias |
| Yosselson-Superstine and Weiss (1982)22 | Israel | Paediatric ward, university hospital | 7 months | Children (0–16 years) | NR | NR | Interview and medical record review | 31/906 (3.42) | Discontinued medication (10/31), UD (11/31), OD (6/31), mixed type (4/31) | 31/160 (19.38) | Infection drugs (64.5), CNS drugs (22.6), CVS drugs (3.2) | Moderate risk of bias |
| Bigby et al (1987)23 | USA | Emergency admissions, teaching hospital | 12 months | General population | 60.7±18.8 years, 36% male | Patient ability to comply with prescribed therapies. | Interview of patients and primary care clinicians | 26/686 (3.79) | NR | 26/73 (35.62) | NR | Low risk of bias |
| Davidsen et al (1988)24 | Denmark | Department of Cardiology, university hospital | 2 months | General population | Non-adhering patients (F: 74.9±9.7 years, M: 73.3±5.7 years), 50% male | A deviation of more than 50% between the dose actually taken and the prescribed drug dose. | Interview and medical record review | 16/426 (3.76) | NR | 16/65 (24.62) | CVS drugs (56.3), respiratory drugs (6.3), other drugs (37.5) | Low risk of bias |
| Grymonpre et al (1988)25 | Canada | Department of Medicine, tertiary hospital | 4 months | Elderly (≥50 years) | 69.8±0.5 years, 57.4% male | A failure to accomplish the goals of treatment because of deliberate non-adherence to a therapeutic programme. | Interview, medical record review and pill count | 26/863 (3.01) | NR | 26/162 (16.05) | NR | Low risk of bias |
| Col et al (1990)2 | USA | Medical ward, community teaching hospital | 3 months | Elderly (≥65 years) | 76.6 years (range 65–99 years), 45.4% male | Any non-trivial deviation from the prescribed medication regimen. | Interview and medical record review | 34/315 (10.79) | UD (81%), OD (17%), misuse (2%), intentional (54%), unintentional (46%) | 34/89 (38.20) | CVS drugs (63.9), respiratory drugs (30.5), endocrine drugs (5.6) | Low risk of bias |
| Stanton et al (1994)26 | Australia | Medical ward, teaching hospital | 10 weeks | General population | Median age 67 years (range 11–97 years), 50.8% male | Deviation from prescribed medication regimen due to non-comprehension, forgetfulness or by choice, producing an exacerbation of symptoms of the patient’s condition. | Interview and medical record review | 10/691 (1.45) | NR | 10/68 (14.71) | Respiratory drugs (60), CVS drugs (40) | Low risk of bias |
| Courtman and Stallings (1995)27 | Canada | Medical ward, tertiary teaching hospital | 139 days | Elderly (≥65 years) | 78 years (range 65–108 years), 41.3% male | NR | Medical record review and drug level analysis | 9/150 (6.00) | NR | 9/46 (19.57) | NR | Low risk of bias |
| Dartnell et al (1996)28 | Australia | Royal Melbourne Hospital, teaching hospital | 1 month | General population | Aged 15–91 years | Patient or carer described drug taking that deviated from prescribed directions; or the patient’s mental condition or home situation together with the presenting condition made non-compliance highly likely; or drug-assay determinations concurred with a doctor’s suspicion of non-compliance. | Interview and medical record review | 15/965 (1.55) | UD (12/15), OD (3/15) | 15/55 (27.27) | CVS drugs (53.3), respiratory drugs (33.3), endocrine (6.7), CNS drugs (6.7) | Low risk of bias |
| Nelson and Talbert (1996)29 | USA | Intensive care unit or internal medicine service, university hospital | 1 month | General population | Median age in drug-related admission=43 years, 60% male | NR | Medical record review and drug level analysis | 48/450 (10.76) | NR | 48/73 (65.75) | NR | Low risk of bias |
| Murad and Chawla (1997)30 | Bahrain | Medical ward, medical centre | 1 month | General population | NR | NR | Medical record review | 206/2167 (9.51) | NR | 206/523 (39.39) | NR | Moderate risk of bias |
| Chan et al (2001)31 | Australia | Medical ward, public acute care hospital | 2 months | Elderly (≥75 years) | 81.8 years (range 75–94 years), 45% male | A deviation from a prescribed medication regimen due to non-comprehension, forgetfulness or by choice, producing exacerbation of symptoms of the patient’s condition. | Interview and medical record review | 9/240 (3.75) | NR | 9/73 (12.33) | NR | Low risk of bias |
| Malhotra et al (2001)32 | India | Medical emergency department, tertiary hospital | 7 months | Elderly (≥65 years) | 72.5±4.7 years (range 65–91 years), 47.1% male | Haynes definition42 as the extent to which the patient’s behaviour coincides with the clinical prescription. | Interview and medical record review | 44/578 (7.61) | UD (71% of all non-compliance), OD (17%), misuse (2%), intentional (63%), unintentional (37%) | 44/83 (53.01) | CVS drugs (61.4), respiratory drugs (18.2), endocrine drugs (11.4), CNS drugs (9.1) | Low risk of bias |
| Martin et al (2002)33 | Spain | Admissions through the emergency department | 9 months | General population | 68.4 years, 58.9% male | Patients did not comply with the prescribed regimen. | Interview and medical record review | 91/1661 (5.48) | NR | 91/215 (42.33) | CVS drugs (48.4), respiratory drugs (24.2), GI drugs (15.4), CNS drugs (4.4), endocrine drugs (3.3), infection drugs (2.2), UTI (1.1), nutrition and blood (1.1) | Low risk of bias |
| Otero López et al (2006)34 | Spain | Medical units, university hospital | 6 months | General population | NR | NR | Interview and medical record review | 19/2643 (0.72) | NR | 42/177 (23.73) | Respiratory, CVS, CNS, endocrine drugs | Low risk of bias |
| Samoy et al (2006)35 | Canada | Internal medicine, teaching hospital | 12 weeks | General population | 69.3±18.8 years, 49.4% male | Any noxious, unintended or undesired effect caused by failure to receive a drug. | Interview and medical record review | 22/565 (3.89) | NR | 22/136 (16.18) | NR | Low risk of bias |
| Kongkaew (2009)36 | UK | Two tertiary hospitals | 18 months | General population | OD group=38.49±16.75 years, 44% male UD group=58±20.8 years, 57.1% male | Haynes definition42 as the extent to which the patient’s behaviour (in terms of taking medication, following diets or executing other lifestyle changes) coincides with the clinical prescription. | Interview and medical record review | 190/3904 (4.87) | OD (148/190), UD (42/190) | 190/604 (31.46) | OD: analgesic, CNS drugs UD: CNS drugs, endocrine | Low risk of bias |
| Singh et al (2011)37 | India | Internal medicine, tertiary hospital | 6 months | General population | 49.8±18.2 years, 60% male | DRP classifications by Hepler and Strand.43 | Interview and medical record review | 55/3560 (1.54) | NR | 55/118 (46.61) | NR | Low risk of bias |
| Al-Arifi et al (2014)38 | Saudi Arabia | Admission via emergency department, tertiary hospital | 1 month | General population | Median age 51 years, 53.33% male | Haynes definition42 as the extent to which the patient’s drug-taking behaviour (in terms of taking medication) coincides with the prescription. | Interview, medical record review and drug level analysis | 17/251 (6.77) | NR | 17/52 (32.69) | NR | Moderate risk of bias |
| Kongkaew (2015)39 | Thailand | Inpatient units of university hospital, general hospital and community hospitals | 16 months | General population | 56.7±17.4 years, 50.1% male | WHO definition1 as the extent to which a person’s behaviour—taking medication, following a diet and/or executing lifestyle changes—corresponds with agreed recommendations from a healthcare provider. | Interview and medical record review | 32/3755 (0.85) | OD (11/32), UD (21/32) | 32/91 (35.16) | NR | Low risk of bias |
| Gustafsson et al (2016)40 | Sweden | Acute internal medicine ward and the orthopaedic ward at university hospital, and county hospital | 3 years | Elderly (≥65 years) | Among drug-related group: mean age 82.4 years, 39.7% male | A deviation from the prescribed medications because of a choice, non-comprehension or forgetfulness, leading to an ADR or exacerbation of symptoms. | Medical record review | 19/458 (4.15) | NR | 20/189 (10.58) | Endocrine, CVS, respiratory, CNS, injection | Low risk of bias |
| Jolivot et al (2016)41 | France | Medical ICU, teaching hospital | 12 months | General population | Median age 65 years, 57.5% male | NR | Interview and medical record review | 31/701 (4.42) | NR | 31/173 (17.92) | NR | Low risk of bias |
CNS, central nervous system; CVS, cardiovascular system; DRP, drug-related problem; GI, gastrointestinal; HA, hospital admission; ICU, intensive care unit; MRP, medication-related problem; NA, non-adherence; NR, not reported; OD, overdosage; UD, underdosage; UTI, urinary tract infection.