Skip to main content

Table 1 Description of studies meeting the inclusion criteria

From: Evidence-based guideline implementation in low and middle income countries: lessons for mental health care

Lead author (year of publication)

Country

Mental or physical health

Study design

Risk of bias assessment

Health condition or behaviour targeted by Clinical Practice Guideline (GPG)

Setting and participants

Brief description of intervention / control conditions and implementation strategies adopted (classified according to EPOC taxonomy)

Primary outcomes

Narrative summary of results

Berwanger et al. [37]

Brazil

Physical health (acute coronary syndromes)

Cluster randomised trial

Low–moderate risk of bias

CPG The intervention included posters containing evidence-based recommendations for the treatment of acute coronary syndromes (ACS)

Setting and participants 36 general public hospitals with emergency centres in major urban areas in Brazil. 19 hospitals were randomised to receive the intervention and 17 to control. The analysis represented 602 patients with a diagnosis of ACS in the intervention group and 548 in the control group. Detail regarding the number and type of health professionals exposed to the intervention is not provided

Intervention A multifaceted educational quality improvement intervention for management of acute coronary syndromes

Implementation strategies adopted were: reminders; case management; educational materials; educational outreach; local opinion leaders

Checklists were also used

Control: routine care

The percentage of eligible patients who received all appropriate evidence-based therapies (aspirin, clopidogrel, anticoagulants and statins) within the first 24 h of admission to hospital without contraindications

A multifaceted educational intervention resulted in statistically significant improvement in the use of evidence-based therapies both within 24 h of admission and at discharge in comparison to routine care

There were no statistically significant differences in the rate of major cardiovascular events, mortality at 30 days, new myocardial infarction or incidence of major bleeding—however, the study was not powered for evaluation of clinical outcomes

Adherence to the reminders and checklists by healthcare professionals in the intervention group was 82.7%

Du et al. [52]

China

Physical health (acute coronary syndromes)

Cluster randomised controlled trial

Moderate-high risk of bias

CPG Clinical pathways for acute coronary syndromes based on the American Heart Association / American College of Cardiology guidelines

Setting and participants 70 urban hospitals throughout China routinely admitting >100 patients annually with suspected acute coronary syndromes were randomly allocated to receive the intervention (n = 32 hospitals, n = 8049 patients) or control (n = 38 hospitals, n = 5731 patients) Detail regarding the number and type of health professionals exposed to the intervention is not provided

Intervention group Hospitals implemented clinical pathways (and accompanying multifaceted implementation strategies) for acute coronary syndromes ‘early’

Implementation strategies adopted: audit and feedback; monitoring the performance of the delivery of healthcare; educational meetings

Control group Usual care for first year. Hospitals implemented clinical pathways (and accompanying implementation strategies) for acute coronary syndromes ‘late’ (1 year after intervention sites)

Key performance indicators in intervention sites were collected 12 months after the introduction of the intervention and compared with baseline data in control sites

The guideline was tailored to the local context. Teams led by senior cardiologists took responsibility for implementation

Proportion of patients with final diagnosis consistent with biomarker findings; pro- portion of patients with ST-segment-elevation myocardial infarction (STEMI) receiving thrombolysis or primary percutaneous coronary intervention (PCI) among those arriving within 12 h of symptom onset; door-to-needle time for patients with STEMI undergoing thrombolysis; door-to-balloon time for patients with STEMI undergoing primary PCI; proportion of high-risk patients undergoing coronary angiography; proportion of low-risk patients (no on-going symptoms, persistently normal electrocardiogram, and persistently normal biomarkers) undergoing functional testing; proportion of patients discharged on combination medical therapy (including any antiplatelet therapy, β-blocker, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and statin); length of stay

The use of a clinical pathway for treatment of acute coronary syndromes compared with usual care statistically significantly improved use of secondary prevention treatments (an increase in the number of patients who were discharged on appropriate medical therapy) but there were no statistically significant differences on other measures of quality of care

A survey of 556 health professionals found that 98.2% had heard of the intervention, and >80% had attended training sessions, used the pathway in clinical practice and were aware of study reports.>90% completely or strongly agreed that the pathway was valuable.

A qualitative process evaluation conducted alongside the trial aids understanding of findings

Okasha et al. [42]

Egypt

Mental health

Uncontrolled before and after study

High risk of bias

CPG The tenth revision of the International Statistical Classification of Mental and Behavioural disorders Primary Health Care (ICD-10 PHC)

Setting and participants: 20 GPs working in 6 primary health care centres (rural and urban)

in Egypt

Intervention: Training was conducted by an experienced psychiatrist. The ICD-10 training kit, an educational training programme and materials produced by the WHO

Implementation strategies therefore comprised: educational meetings and educational materials

GP’s attitudes, knowledge, interview skills and ability for diagnosing psychiatric disorders in a primary care setting

There was statistically significant improvement in GP’s optimism about helping patients with mental disorders and in their confidence in their ability to diagnose mental disorders following the intervention. There was no statistically significant change in GP’s interest in patients with mental disorders or in their reporting of the importance of diagnosing mental disorders.

Pagaiya and Garner [41]

Thailand

Physical health (diabetes mellitus) and mental health (anxiety and panic disorder)

Cluster randomised controlled trial

Moderate risk of bias

CPG Four clinical guidelines introduced: two for children (acute respiratory infection and diarrhoea); two for adults (diazepam prescribing for anxiety and panic disorder and management of diabetes mellitus)

Setting and participants Health centres in Khon Kaen province in Thailand, mainly in rural areas. 9 health centres were randomised to receive the intervention and 9 to control. 110 patients for diazepam prescribing and all diabetes patients at each centre were included

Intervention group introduction of four clinical guidelines (two of which meet the inclusion criteria for the present review) plus the following implementation strategies: educational meetings; educational out-reach visits; audit and feedback

Control group usual care.

The PRECEDE-PROCEED implementation framework was adopted

Diazepam prescribing; prescribing costs per patient; management of diabetes

Clinical guidelines implemented with educational meetings and outreach visits and audit and feedback improved some aspects of prescribing but not others in the short-term

Diazepam prescribing was statistically significantly reduced in the intervention group.

There was no statistically significant difference in the management of diabetes between the intervention and control groups.

For all four CPGs combined, average drug costs per patient in the intervention group were statistically significantly less than in the control group

Shrestha et al. [39]

Nepal

Physical health (asthma, COPD)

Cluster randomised trial

High risk of bias

CPG Practical Approach to Lung health (PAL) guidelines on prescription behaviour and total cost of prescription for patients with asthma, pneumonia and chronic obstructive pulmonary disease (COPD)

Setting and participants: Of 76 health facilities in Nawalparasi district, 40 were included in the study on the basis of highest patient attendance. Facilities were stratified by type and subsequently randomized to receive the intervention [21] or usual practice control [19]. Detail regarding the number of patients in each group is not provided, rather number of prescriptions is given. Detail regarding the number and type of health professionals exposed to the intervention is not provided

Intervention Introduction of the PAL guideline plus the following accompanying implementation strategies: educational materials; educational meetings; educational outreach (train-the trainer)

Control Usual practice

The guideline was adapted to the Nepalese context. Different professionals, organisations and health workers were involved in this process

Healthcare professional prescribing practices for asthma and COPD

The guidelines and accompanying implementation strategies led to a statistically significant reduction in poly-pharmacy (multiple prescriptions) in comparison to the control condition. There were no statistically significant differences in the prescription of generic and essential drugs, average prescription cost and wastage cost, or prescription of antibiotics between the intervention and control groups

Steyn et al. [40]

South Africa

Physical health (diabetes mellitus and hypertension)

Cluster randomised controlled trial

Low -moderate risk of bias

CPG A structured record which incorporated the national guidelines for the management of patients with type 2

Diabetes mellitus or hypertension or both conditions

Setting and participants 18 randomly selected public sector primary healthcare clinics known as Community Health Centres (CHCs) in Cape Town in 1999 and 2000. 9 clinics were randomised to deliver the intervention and 9 to control. At baseline, in the intervention clusters there were n = 229 patients with diabetes and n = 461 patients with hypertension. At baseline, in the control clusters there were n = 227 patients with diabetes and n = 459 patients with hypertension. The intervention was targeted at General Practitioners and nurses who consult patients with chronic disorders. Detail regarding the number of health professionals exposed to the intervention is not provided

Intervention a structured record that included algorithms for diagnosis and guideline-based management of diabetes and hypertension, including educational topics to be covered with the patient

Implementation strategies adopted: case management; reminders; educational outreach from a local diabetes and hypertension expert

Control group usual care, which included the guidelines passively disseminated by the National Department of Health

Mean level of glycated haemoglobin in all patients with diabetes

Mean systolic and mean diastolic BP in all those with hypertension as measured at the end of the 1 year intervention period

The intervention had no statistically significant impact on either diabetes or hypertension control

No differences were observed in the recording of process measures, e.g. visual acuity or foot examinations

The process measures collected implied that the structured record was not widely adopted by healthcare professionals in the primary care clinics. At follow-up, the structured record was found in the folders of only 58% of patients with diabetes and 47% of patients with hypertension