A well-formulated clinical question has to be focused on specific issues: which aspect do we want to address?: a particular therapy; preventive effects influencing quality of life; economic feasibility of a procedure.

The clinical question can be reduced to four types:

- Therapy.

- Prognosis

- Diagnosis.

- Economic analysis/decision.

A good question (quantitative effect-focused studies) usually incorporates the PICO elements:

Population
Intervention
Comparison 
Outcome 

Other variants such as PICOS or PICOT incorporate the study design (S) or time frame (T). PICo is used in qualitative (experience-focused) research:

Population
Phenomenon of Interest
Context

This PICO model guides us in the process of constructing the research strategy.

To ensure the scientific precision of your review paper you will need to consider two aspects: the inclusion and exclusion criteria, and the methodology.

The inclusion criteria may be affected by the available studies and at the same time determine which research articles will be selected.

Establishing the inclusion criteria requires the following issues to be addressed:

• Typology of studies to be included according to their design.
• Intervention, activity or phenomenon of interest on which the study focuses.
• Outcome (especially in quantitative research). For example, the efficacy of a drug or intervention.
• Study population (e.g., age groups or groups with specific characteristics).
• Languages of publication.
• Chronological period.

The methodology follows this sequence: literature search >>evaluation of retrieved studies>>extraction and synthesis of data>>results>>conclusions>>implications for clinical practice.

Construction of the search strategy and the search for evidence.

The objective is to identify as many studies on the topic of interest as is feasible in a comprehensive search. If you don't find the evidence it cannot be reviewed.

Conducting a comprehensive search requires a number of processes:

- Developing a search string.

- Searching bibliographic databases that index health sciences research.

- Searching grey literature (unpublished).

- Manual search.

Searching is an interactive process, involving continuous evaluation and filtering of results. The usefulness of search terms, descriptors and keywords is determined by the search result.

The search has to be both sensitive and specific to retrieve relevant studies and exclude irrelevant ones.

Developing the search strategy: it is essential to define the review question and the inclusion criteria.

The following issues need to be addressed:

- Type of data (quantitative, qualitative, economic).

- Type of studies (RCTs, cohort studies, ethnographic studies, etc.).

- Limits and restrictions of the research (date of publication, language).

Search example:

"Is animal-assisted therapy more effective than music therapy in managing aggressive behaviour in elderly people with dementia?"

POPULATION: elderly patients with dementia.

INTERVENTION: animal-assisted therapy.

COMPARISON INTERVENTION: music therapy.

OUTCOME MEASURES: aggressive behaviour.

Steps to follow:

1. Keywords and free-text words, synonyms: the title and abstract of the retrieved articles must be examined to locate the key terms used in the literature (pet therapy).
2. Indexed terms and subject headings (thesaurus). This is a controlled vocabulary with a list of standardised terms (MeSH ; EMTree).
3. Make a grid with PICO in the columns and fill them with the indexing terms, keywords and subject headings.
4. The structure of the search strategy remains the same regardless of the database, but the indexing terms need to be adapted to each database.

Where to search?

Indispensable databases in a systematic review:

1. MEDLINE
2. EMBASE
3. COCHRANE
4. CINAHL; PSYCINFO (depending on the subject)
5. Other important databases (Scopus, Web of Science, Google Scholar).
6. Manual search.

Search grey literature:

1. Websites of relevant organisations.
2. Grey literature databases (GreyNet International, Grey Literature Report).
3. Search engines.

Report the search strategy:

The strategy, including the filters and databases, should be communicated. Many researchers use the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines.

There are two phases in this process:
1st: Reviewing the documents retrieved to see which ones we want to read in full text (Screening).
2nd: Critical appraisal of those documents selected for reading.
Critical appraisal differs whether the studies are quantitative (measuring the effectiveness of an intervention) or qualitative (examining individual meaning and experience).

PRISMA Flow Diagram Link

Inclusion and exclusion criteria need to be considered in relation to:
- The chronological period.
- The language of the articles.
- See if the study population fits these inclusion/exclusion criteria.
- Is the study design reported and is it relevant to the review?
- Is there an outcome measure?

INTERNAL VALIDITY OF STUDIES: assesses the quality of studies in terms of the causal relationship between the intervention and the outcomes, i.e. how well these outcomes can be inferred from the findings.

Tools for critical appraisal: JOANNA BRIGGS INSTITUTE Checklist

This phase of the review process is carried out by a health sciences researcher and goes beyond the field of information and documentation.

We recommend the advice of academic staff.

The standard for the presentation of results is:

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).

The following elements should be documented:
1. The selection of studies. Including the PRISMA flow diagram with specific details of why studies were included or excluded. Databases used, manual searches and risks of bias (selection bias, publication bias).

2. Assignment of levels of evidence to recommendations. In the summary of findings, the level of evidence must be graded for each of the outcomes identified in the review protocol.
From the risk/benefit assessment we obtain strong vs. weak recommendations.

3. Implications for practice. The following should be assessed:

• Evidence of feasibility.
• Evidence of appropriateness.
• Evidence of meaningfulness.
• Evidence of effectiveness.

4. Statistical vs. clinical significance. Statistical significance does not always translate into relevance for clinical practice.

• Risk/hazard ratio.

5. Implications for further research

6. Limitations and conclusions

MEDLINE via OVID - Search strategy:

1. Oral Health/
2. Dental Caries/
3. Dental Care/
4. Dental Care for Children/
5. Periodontal Diseases/
6. Gingival Diseases/
7. Oral Hygiene/
8. Periodontitis/
9. ((oral adj2 care) or (mouth adj2 care) or (dental adj2 care) or (teeth adj2 care)).mp.
10. ((oral adj2 health) or (dental adj2 health) or (teeth adj2 health) or (mouth adj2 health)).mp.
11. ((oral adj2 hygiene) or (mouth adj2 hygiene) or (dental adj2 hygiene) or (teeth adj2 hygiene)).mp.
12. (dental caries or dental decay).mp.
13. (gingival diseas* or periodont*).mp.
14. or/1-13 

Documents related to dental care and hygiene are retrieved.
In line 14 all documents are added together to retrieve them all together (OR operator).

     15. Preventive Dentistry/

     16. Health Promotion/

     17. Health Education, Dental/

     18. (health adj2 (promot* or advi?e or educat* or practic* or improv*)).mp.

     19. (prevent adj2 oral).mp.

     20. (prevent adj2 dent*).mp.

     21. 0r/15-20

Documents related to prevention and oral health education are retrieved.
In line 21 all documents with the operator OR are added together

     22. Midwifery/

     23. Nurses/

     24. Education, Nursing/ or Nursing/ or Public Health Nursing/ or Neonatal Nursing/ or Pediatric Nursing/ or Maternal-Child Nursing/ or Community Health Nursing/ or Primary Care Nursing/ or Obstetric Nursing/ or Nursing Staff/

    25. Health Personnel/

    26. ((nurs* adj2 educat*) or (midwi* adj2 educat*) or nurs*adj2 train or (midwi* adj2 train*) or (nurs* adj2 practic*) or (midwi* adj2 practic*)).mp.

    27. (health visit* or home visit*).mp.

    28. Or/22-27

Nurses and Midwives

29. 14 and 21 and 28

(We retrieve documents that meet all search conditions)

DENTAL CARE AND PREVENTION/PROMOTION AND NURSES/MIDWIVES