Interventions for improving coverage of child immunization in low‐income and middle‐income countries
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
The aim of this review is to evaluate the effectiveness of intervention strategies to boost and sustain high childhood immunization coverage in LMIC.
Background
The immune system can be stimulated by the introduction of killed, or live‐attenuated agent, or modified toxins into the body. This process is termed immunization and can be directed against specific disease causing organisms. By this many diseases can be controlled and sometimes eradicated. Immunization is reported to be second to clean water in reducing the burden of infectious diseases (Andre 2008). Vaccines are available for the following diseases: tuberculosis, diphtheria, measles, tetanus, Hepatitis B infection, poliomyelitis, Haemophilus Influenza, pertussis, yellow fever, mumps, rubella, pneumococcal, rotavirus, cholera, etc. Immunization is said to be the single most efficient and cost effective means of controlling these diseases (NSW 2003; JAMA 2006). This is evident in the drastic decline and in some cases, elimination, of certain infectious diseases since the introduction of vaccines in the 20th century (CDC 1999a; NSW 2003).
Vaccines are not only used in preventing diseases, they are useful in the mitigation of the severity of disease, prevention of infections, prevention of cancers (e.g. cancer of the cervix, cancer of the liver), and reduction in the complications associated with the infections (Andre 2008). When a sufficient proportion of the population is immune there is an indirect effect on the whole population called "herd immunity" (Andre 2008). This causes a reduction in the spread of the infective agent of the diseases by blocking the transmission from one person to another.
Description of the condition
There have been concerted efforts by the World Health Organization (WHO) to boost immunization coverage globally. One of such efforts was the launching of the Expanded Program on Immunization (EPI) in 1974. The target of the program was to achieve 80% coverage of children aged less than 2 years with vaccines against 6 childhood killer diseases namely measles, diphtheria, tetanus, polio, Bacille Calmette‐Guérin (BCG), and pertussis vaccines by 1990 (Piotrow 1992; Worldbank 2009). By 1980 the coverage of DPT3 was estimated to be only 20%. By 2003, however, coverage had increased to 78% globally. The progress in low and middle‐income countries (LMIC), gross national income per capita of $11,455 or less (Worldbank 2009), is slow nevertheless and DPT3 coverage in sub‐Saharan Africa is estimated to be 60%. Of the estimated 27 million children that were yet to be reached with DPT3 vaccine, 9.9 million were in South Asia and 9.6 million in sub‐Saharan Africa (WHO 2004). In 2002, WHO had estimated that about 1.9 million of the 2.5 million (76%) vaccine preventable deaths among children aged less than 5 years worldwide occurred in Africa or South East Asia (JAMA 2006). Among these childhood deaths, over 500,000 were caused by measles; nearly 400,000 by Haemophilus influenza B; 300,000 by pertussis; and 180,000 by neonatal tetanus (WHO 2004).
This trend leaves the attainment of the millennium development goal of reducing child mortality rate by two thirds by 2015 elusive, particularly in LMIC. So far, only 16% of LMIC countries are on track to achieving this goal, and none are in sub‐Saharan Africa (GAVI 2005). Poor coverage has been attributed to poor access to vaccines and high dropout rates (Nath 2007). A strategy for immunization that will achieve a high and sustainable coverage is needed in LMIC countries.
Description of the intervention
Currently vaccines are made accessible through routine immunization provided in fixed facilities (such as health centres, outpatient clinics, and district hospitals), mobile strategies, immunization outreach programs, extended outreach programs, and immunization campaigns. In fixed‐facility strategy, vaccines are provided on a routine basis in static health facilities at different levels of the health system. National or sub‐national immunization campaigns can be carried out for specific vaccines, and these are usually targeted at boosting ongoing immunization activities. For mobile strategy immunization, specialized vehicles are used to convey vaccines to remote areas. When health workers convey vaccines from the health facility to the homes of the people in the community this is termed outreach program. When the outreach is intensive (e.g. reaching out to target population at homes, markets, places of worship, at remote communities) it is known as "extensive outreach" (Brenzel 2006).
How the intervention might work
Strategies for improving immunization coverage could be patient‐oriented interventions, provider‐oriented interventions, or system interventions (Jacobson Vann 2005). Patient‐oriented interventions aim at increasing demand for vaccination by the patient, e.g. patient reminder recall or health education of clients. Provider‐oriented intervention aims at reducing missed opportunities, such as audit and feedback and chart‐based or computerized provider reminders. System interventions improve access to the services through such methods as outreach programs and improve quality of delivery of care (CDC 1999b). In a Cochrane systematic review a patient‐oriented intervention, patient reminder and recall, was reviewed. The evidence indicated that reminding people to receive immunization through postcards, letters or telephone calls increased immunization uptake (NSW 2003). This strategy generally relies on setting up an efficient computerized immunization registry or other practice‐based tickler systems to track clients' immunization status and eligibility for recommended vaccines and also an efficient communication system to send reminders to clients. These technologies are lacking in LMIC.
Why it is important to do this review
This review will examine the effects of strategies for improving immunization coverage in LMIC. Such strategies have the potential to boost coverage, and indirectly reduce disease burden.
Objectives
The aim of this review is to evaluate the effectiveness of intervention strategies to boost and sustain high childhood immunization coverage in LMIC.
Methods
Criteria for considering studies for this review
Types of studies
1. Randomized controlled trials (RCT)
2. Non‐randomized clinical trials (NRCT)
3. Interrupted time series (ITS) (with clearly defined point at which intervention occurred and at least three data points before and three after the intervention)
Types of participants
Community or institutional based studies in LMIC that include:
-
children aged 0‐4 (under 5) years who receive globally recommended vaccines which include any of the following: diphtheria, pertussis, tetanus, measles, mumps, rubella, (as single or combined antigens), polio, BCG Hepatitis B, Haemophilus Influenza.
Types of interventions
Interventions:
1. Patient‐ or community‐oriented interventions, e.g.
-
Vaccination requirement for school entry
-
Client incentives
-
Health education
2. Provider‐oriented interventions, e.g.
-
Any intervention to reduce missed opportunity (e.g. audit and feedback; provider reminders; fact sheet provider reminders)
-
Health education/ training/update course for provider
3. Health System intervention, e.g.
-
Interventions to improve quality of services such as provision of reliable cold chain system, provision of transport for vaccination, vaccine stock management, etc)
-
Outreach program e.g. school immunization outreach program, door‐to‐door canvassing (channeling), immunization campaigns (national and sub‐national)
-
Expanded services e.g. extended hours of immunization
-
Budget for immunization
-
Integration of immunization services with other services
-
Plan of action for immunization coverage and disease reduction goals
4. Multi‐faceted (any combination of the above categories of interventions)
5. Single or multiple intervention other than the above
Exclusion:
Patient reminder recall (to be excluded because of the existing review on this)
Comparisons:
1. Routine immunization practices in the study setting.
2. Different interventions or different degrees of intensity.
Types of outcome measures
Primary outcomes
1. Proportion of target population fully immunized by age with recommended vaccines.
2. Number of children aged two years fully immunized per vaccine
Secondary outcomes
1. Occurrence of vaccine preventable diseases
2. Number of under‐fives fully immunized with all scheduled vaccines
3. Number of under‐fives partially immunized for multi‐dose vaccines
4. Cost of intervention
5. Attitude of caregivers/clients towards immunization
6. Unintended adverse effects
Search methods for identification of studies
See: Effective Practice and Organisation of Care Group methods used in reviews.
Electronic searches
We will search the Database of Abstracts of Reviews of Effects (DARE) for related reviews.
We will search the following electronic databases for primary studies:
-
Cochrane Effective Practice and Organisation of Care Group Specialized Register (and the database of studies awaiting assessment)
-
Cochrane Central Register of Controlled Trials (CENTRAL)
-
MEDLINE
-
EMBASE
-
CINAHL
-
LILACS
-
Sociological Abstracts
We will develop strategies that incorporate the methodological component of the EPOC search strategy combined with selected index terms and free text terms. The MEDLINE search strategy will be translated into the other databases using the appropriate controlled vocabulary as applicable.
See Appendix 1 for MEDLINE, OVID search strategy.
Searching other resources
-
Reference lists of all papers and relevant reviews identified
-
We will contact authors of relevant papers regarding any further published or unpublished work
-
We will search ISI Web of Science for papers that cite studies included in the review
Data collection and analysis
Selection of studies
CO will screen titles and abstracts of studies for inclusion and CN will retrieve the full text of potentially eligible studies for screening and will independently apply the inclusion criteria to the retrieved publications. We will discuss any disagreements about the inclusion of studies and if no consensus is met, we will consult the contact editor for the review. We will seek further information from the authors where papers contain insufficient information to make a decision about eligibility.
Data extraction and management
AO will develop a data extraction form to be reviewed by all authors. CN will perform data extraction and risk of bias assessment. AO will check and enter the data into RevMan 5. Any disagreement in data extraction will be resolved by MM or by consensus between the four authors. Should there be missing or inadequate data, MM will attempt to obtain such by contacting the authors.
Assessment of risk of bias in included studies
We will use the EPOC nine point criteria for randomised controlled trials (RCTs), and non‐randomized clinical trials (NRCT), and seven point criteria for interrupted time series (ITS) studies to determine the quality of all eligible studies. AO and CN will apply the criteria, which are given below. Disagreement will be discussed with MM. If we fail to resolve the disagreement, we will seek the opinion of the contact editor.
Criteria for randomised controlled and non‐randomized controlled trials
-
Was the allocation sequence adequately generated?
-
Was the allocation adequately concealed?
-
Were baseline outcome measurements similar?
-
Were baseline characteristics similar?
-
Were incomplete outcome data adequately addressed?
-
Was knowledge of the allocated interventions adequately prevented during the study?
-
Was the study adequately protected against contamination?
-
Was the study free from selective outcome reporting?
-
Was the study free from other risks of bias?
Criteria for interrupted time series (ITS) studies
-
Was the intervention independent of other changes?
-
Was the shape of the intervention effect pre‐specified?
-
Was the intervention unlikely to affect data collection?
-
Was knowledge of the allocated interventions adequately prevented during the study?
-
Were incomplete outcome data adequately addressed?
-
Was the study free from selective outcome reporting?
-
Was the study free from other risks of bias?
Each criteria shall be scored as "YES", "NO" or "NOT CLEAR". Risk of bias for outcomes shall be summarized across studies as follows:
Low risk of bias = Most information is from studies at low risk of bias
High risk of bias = The proportion of information from studies at high risk of bias is sufficient to affect the interpretation of results
Unclear risk of bias = Most information is from studies at low or unclear risk of bias
AO will analyse the data and write up all sections of the review. MM will respond to comments from editors and finalize the writing of the review.
Measures of treatment effect
We will use risk ratio in analyses of dichotomous data on desirable outcome (i.e. participants vaccinated). The weighted mean difference will be calculated for costs and any other analyses of continuous data. The random effects model will be used as the default procedure in meta‐analysis.
When possible, the results of our analyses of the effects of interventions on immunization coverage will be interpreted in the context of systematic reviews of the effects of an intervention across different outcomes. For example, the effects of client incentives on immunization coverage will be interpreted in the context of systematic reviews of the effects of client incentives more broadly (Giuffrida 1997; Kane 2004; Lagarde 2007).
Unit of analysis issues
Cluster‐randomized trials will be included in meta‐analysis only if adjustments were made for design effect. Where possible, we will correct for design effects using standard procedures (Rao 1992). We will calculate design effect for each trial using the following formula: design effect = 1+(m ‐ 1)r, where m is the average cluster size and r is the intra‐cluster correlation coefficient.
Dealing with missing data
We will contact study authors to provide additional data wherever there are missing data. See sensitivity analysis section below.
Assessment of heterogeneity
Heterogeneity will be considered "statistically significant" if the p value for the Chi2 test is <0.1 or the I statistic is 50% or more. Where we find significant heterogeneity and decide not to pool data, we will present such data in additional tables.
Assessment of reporting biases
The risk of publication bias will be investigated using a funnel plot across interventions. If there are ten (10) or more studies in a meta‐analysis, we will visually examine the plot for asymmetry and seek the help of a statistician.
Data synthesis
Data from studies of similar interventions (i.e. with similar participants, outcomes and study designs) will be pooled in meta‐analysis unless there is significant statistical heterogeneity or methodological differences, or high risk of bias. For ITS studies we will report change in level and change in slope.
Subgroup analysis and investigation of heterogeneity
We will test the following hypotheses:
-
Larger incentives shall produce a higher proportion of children immunized
-
High risk of bias shall produce a larger effect size
-
One dose vaccine shall have a higher proportion of coverage than multiple dose vaccine
We will use sub‐group analysis to explore these differences.
Sensitivity analysis
If there are sufficient data we will perform sensitivity analysis based on risk of bias and missing data.
