Skip to main content

A protocol for a systematic literature review: comparing the impact of seasonal and meteorological parameters on acute respiratory infections in Indigenous and non-Indigenous peoples



Acute respiratory infections (ARI) are a leading cause of morbidity and mortality globally, and are often linked to seasonal and/or meteorological conditions. Globally, Indigenous peoples may experience a different burden of ARI compared to non-Indigenous peoples. This protocol outlines our process for conducting a systematic review to investigate whether associations between ARI and seasonal or meteorological parameters differ between Indigenous and non-Indigenous groups residing in the same geographical region.


A search string will be used to search PubMed®, CAB Abstracts/CAB Direct©, and Science Citation Index® aggregator databases. Articles will be screened using inclusion/exclusion criteria applied first at the title and abstract level, and then at the full article level by two independent reviewers. Articles maintained after full article screening will undergo risk of bias assessment and data will be extracted. Heterogeneity tests, meta-analysis, and forest and funnel plots will be used to synthesize the results of eligible studies.

Discussion and registration

This protocol paper describes our systematic review methods to identify and analyze relevant ARI, season, and meteorological literature with robust reporting. The results are intended to improve our understanding of potential associations between seasonal and meteorological parameters and ARI and, if identified, whether this association varies by place, population, or other characteristics. The protocol is registered in the PROSPERO database (#38051).

Peer Review reports


Acute respiratory infections (ARI) contribute to a substantial global burden of morbidity and mortality [13]. An estimated 14.9 million children were hospitalized for ARI in 2010, of which 265,000 died [3]. Defined as an acute infection with coughing as a symptom [4], ARI is often associated with meteorological parameters [1] and commonly varies by season [5]. For instance, ARI is commonly associated with temperature parameters, with increasing incidence during cold periods as a result of an individual’s exposure, susceptibility, and the infection type [6]. Furthermore, seasonal associations with ARI also have been identified; in some cases, these seasonal associations have been attributed to varying meteorological parameters and in other cases attributed to the pathogen’s own rhythmicity [6]. The associations between ARI and meteorological parameters and season may be modified by social gradients of health [79] and affected by type of livelihood [8, 10]. For instance, among Indigenous peoples, a strong connection to the land [1113], resource-based livelihoods [14, 15], interacting social determinants of health (i.e., housing, education) [9, 14, 16], and the legacies of colonization [9] may modify the association between infections and meteorological and seasonal parameters [14]. Therefore, it is possible that there are differences in the association between weather variables and ARI among Indigenous and non-Indigenous communities. It is important to know if these associations differ between Indigenous and non-Indigenous communities to aid in better planning, resource allocation, and intervention strategies. This paper outlines a protocol for conducting a systematic review to investigate whether associations between ARI and seasonal or meteorological parameters differ between Indigenous and non-Indigenous groups residing in the same geographical region.

Methods and design

This protocol, which outlines methods for the proposed systematic review, was designed in accordance with the Preferred Reporting Items for Systematic review and Meta-Analyses (PRISMA) Guidelines [17]. The protocol is registered in the PROSPERO database (#38051). The items of this protocol are presented in accordance with the PRISMA-P checklist (Additional file 1) [18].

Review question

This systematic review protocol outlines the procedures for a systematic literature review that is intended to answer the question: Is the association between seasonal or meteorological parameters and ARI the same in Indigenous and non-Indigenous peoples who live in the same geographical region?

The components of population, exposure, comparator, and outcome (PECO) are as follows:

  • Population: communities with Indigenous and non-Indigenous community members;

  • Exposure: indigeneity;

  • Comparator: non-Indigenous; and

  • Outcome: association between seasonal or meteorological parameters and ARI.

Study designs eligible

All primary epidemiological observational study designs (i.e., cross-sectional, cohort, case-control studies) are eligible for inclusion (Table 1). Ecological studies will be eligible, as the population in a singular location should be equally exposed to seasonal or meteorological parameters. Experimental studies (i.e., intervention studies) will not be eligible, as the exposure (i.e., indigeneity) cannot be assigned. Further, reviews, commentaries, editorials, mathematical models, or other non-primary research articles will not be eligible, as these studies are not comparable with observational results.

Table 1 Inclusion and exclusion criteria for a systematic literature review investigating the impact of seasonal and meteorological parameters on acute respiratory infection (ARI) in Indigenous and non-Indigenous peoples

Participants eligible

Eligible study populations are those in which a portion of the population living in the same region is explicitly defined as Indigenous and a portion of the population is explicitly defined as non-Indigenous. This research builds from the United Nations Declaration of the Rights of Indigenous Peoples [13] understanding of the term Indigenous peoples, which states that an Indigenous person self-identifies as Indigenous; has historical continuity with pre-colonial society; has a strong link to territory and natural resources; has a distinct social, economic, or political system; has a distinct language, culture, and/or belief system; forms a non-dominant societal group; and/or resolves to maintain and reproduce their ancestral environments and systems as distinctive peoples and societies.

Exposures eligible

Indigeneity is the only eligible exposure. The comparator group is a non-Indigenous population living in the same region as the Indigenous population.

Outcome measures eligible

All eligible studies will present the association between ARI and weather variables. Eligible studies must investigate this association among both Indigenous and non-Indigenous peoples living in the same region. The results can be presented in one of two eligible ways: (i) two models representing the Indigenous and non-Indigenous strata separately with the same seasonal or meteorological parameter in both models, producing two measures of association for which the ratio of odds ratios (ROR) can be calculated (i.e., ARI-exposure model for Indigenous peoples and ARI-exposure model for non-Indigenous peoples); or (ii) a model for an ARI outcome that includes indigeneity and seasonal or meteorological parameter(s) as independent variables from which further calculations can be made (Fig. 1).

Fig. 1
figure 1

Options for presentation of results in eligible studies, where Option 1 represents a study presenting two models and Option 2 represents a study presenting one model. E+ study exposure positive, E- study exposure negative, SE standard error, β coefficient

The case definition for ARI is an acute infection (i.e., less than 14 days duration, if duration is stated) with coughing as a primary symptom, with or without any accompanying symptoms. If duration is not provided in the case definition, the word “acute” will be sufficient for inclusion. The majority of ARI outcomes are anticipated to be respiratory syncytial virus, influenza, or pneumonia; however, studies reporting any ARI outcome are eligible. A diagnosis of ARI may include any symptomatic description that meets the case definition, a positive biological sample (e.g., swab), clinical diagnosis from a practitioner, or self-reported illness. Studies of non-infectious respiratory outcomes (e.g., asthma) are not eligible.

Seasonal variables refers to a pattern (i.e., season) in meteorological parameters related to a predictable trend such as temperature, hours of sunlight, or total precipitation that is repeated annually [19]. Meteorological parameters are defined as observable weather events, at a point in time, primarily consisting of temperature, precipitation, barometric pressure, humidity, sunlight, and the interactions and variability of these parameters [20]. In this review, any study that has seasonal or meteorological parameters measured at two or more points in time, and where the points are meteorologically or seasonally contrasting (i.e., measured in two different dates or in two different seasons), will be eligible.

Search methods for the identification of studies

The search strategy comprises three main components: Indigenous communities (population and exposure terms); and association between seasonal or meteorological parameters and ARI (outcome terms; Table 2). Terms that will be used to identify Indigenous peoples globally are based on a series of umbrella terms for Indigenous used globally and throughout time, adapted from Bartlett et al. [21]. Individual group names were added to the umbrella terms from two sources. First, the International Work Group for Indigenous Affairs (IWGIA; registry provides a continental directory of Indigenous peoples, further sorted as a country-by-country list of recognized Indigenous groups. These terms may be at the greater population level, rather than individual group level if the category can be expected to represent and include all of the unique peoples groups within it (i.e., the name “Maori” was included in the search terms, but individual Maori group names, such as Ngati Kuri, Ngati Maru, and To Arawa, were not included). Secondly, the United Nations Refugee Agency (UNHCR) provides a country-by-country database of minority and Indigenous peoples ( Since this list provides both Indigenous and non-Indigenous peoples, only those groups explicitly listed as Indigenous were collected. The names of all identified groups defined as Indigenous were added to the search. When the two lists were complete, the lists were merged into one, alphabetized, and de-duplicated. The list of terms for Indigenous peoples is comprehensive to the best of the ability of this strategy. Searches of MeSH terms for “season,” “meteorology,” and “weather” were performed in PubMed® and used to compile terms for the search strategy. Terms that will be used to identify ARI outcomes in the literature include any pathogen known to primarily cause ARI (i.e., enteric pathogens that rarely cause ARI are not included) based on the Medical Microbiology 4th Edition, Chapter 93, online version [22]. Any terms used for ARI by the Lung Disease Alphabetical Listing generated by the American Lung Association ( were added to the search strategy. A search of MeSH terms for “respiratory” and “lung infection” was performed in PubMed® and any additional terms were added to the search strategy.

Table 2 Search string prepared for PubMed®

It is difficult to develop a search strategy that is robust enough to represent all nuances of the terms Indigenous, seasonal or meteorological parameters, and ARI, and thus, the use of multiple databases is intended to increase sensitivity of the search. This review will search the following databases: PubMed® (via OvidSP®), CAB Abstracts/CAB Direct©, and Science Citation Index® (via Web of Knowledge™). The search string will be appropriately adapted for each of the selected databases (Tables 2, 3, and 4). A university librarian was consulted in preparation of the search strategy for PubMed®.

Table 3 Search string prepared for Web of Knowledge™
Table 4 Search string prepared for CAB Abstracts/CAB Direct©

The search will not be limited by language, date, or study design. Search terms will be in English, although the names of Indigenous groups are commonly stated in their own languages (i.e., non-English names in the roman alphabet syntax will be used). An English search string should identify all English articles and any non-English articles with an English title and abstract. If a non-English citation is collected by the search, Google Translate© will be used to translate the title and abstract for initial screening [23], and if maintained for full article screening, Google Translate© will be used for full text screening. This will allow calculation of the total number of eligible articles to generate appropriate denominators. If after full article screening a non-English article is eligible, the article will be formally translated by a paid service, if funding is available. If it is not possible to translate non-English articles, they will be excluded from data extraction and risk of bias (i.e., after full-text screening).

To minimize the risk of exclusion of relevant citations, the citation list of each included study will be searched (i.e., a snowball search). Additionally, Google Scholar© will be used to complete a citation search on eligible studies, to identify studies that have referenced these studies. Studies identified by either the hand-searches or citation searches will be screened for relevance.

Published and unpublished literature will be eligible. Published literature can be collected by all of the proposed databases. Unpublished literature can be collected by CAB Abstract/CAB Direct© and Science Citation Index®. Unpublished primary research that exceeds 500 words in length will be eligible if it meets one of the three following criteria: (i) governmental report (i.e., produced by a regional or national government ministry); (ii) non-governmental report (i.e., produced by non-governmental organizations); or (iii) graduate or honor undergraduate thesis or dissertation. Reports that are fewer than 500 words will be excluded.

If the initial search identifies any relevant government or non-governmental reports, or theses, a search will be conducted in PubMed® to identify any relevant journal publications by the first author of the citation. This PubMed® search will include the first author’s last name and first initial, institutional affiliation, and one to three keywords from the abstract (i.e., author identified key words if available, or reviewer key words if not available).

To keep the review current, if more than 12 months pass from the date the search was conducted to completion of data extraction and analysis, an update search will be conducted. If conducted, the second search will use the same search strategy as the first and will not be restricted by date. The search will be conducted in all of the original databases. Thus, a recall strategy is employed [24], which should identify all of the initial studies and all studies published since the previous search. A hand-matching method will be used to identify whether all of the original citations are included.

Selection of eligible studies

All search result citations will be loaded into and managed in EndNote™ bibliographic software and de-duplicated automatically. Then, citations will be uploaded from EndNote™ into DistillerSR®, which will be used for form generation, screening, and management of relevant screening level statistics.

Screening will be completed in two stages. Screening processes will be piloted and tested by the reviewers on a subset of studies (5% of studies if n > 50, 10% of studies if n ≤ 50). First, title and abstract screening will be conducted on all citations identified. Two reviewers with graduate-level training in epidemiology and systematic literature review processes will screen articles independently, using five evaluation questions (Table 5). All questions will be answered as “yes,” “no,” or “unsure.” In this screening phase, questions will be hidden. A hidden question will not be answered if the article is excluded based on previous screening questions. Articles will be excluded if both reviewers answer “no” to any of the five questions. If both reviewers answer “yes” and/or “unsure” to all questions it will be maintained for full article screening. Any disagreements will be resolved by consensus. When consensus cannot be reached, a third reviewer will arbitrate.

Table 5 Title and abstract screening questions to be used to identify literature for inclusion in the full article screening process

Second, full article screening will be conducted on all citations remaining after title and abstract screening. Two reviewers will screen articles independently, using a second form in DistillerSR®. Reviewers will use seven evaluation questions (Table 6). In full article screening, reviewers will identify any studies using a duplicate dataset and will maintain the research that is most comprehensive. Duplicate results will be removed. Questions will not be hidden in full article screening, allowing for analysis of the reason for exclusion. Studies will be excluded if both reviewers answer “no” to any question. Disagreements between reviewers will be resolved by consensus and, when consensus cannot be reached, a third reviewer will arbitrate.

Table 6 Full article screening questions to be used to identify literature for inclusion

Data collection from eligible studies

Data extraction will include study identifiers and study design; participant, exposure, and outcome information; and information about analytical methods. Missing information will be noted.

Extracted study identifiers will be the authors’ names; study title; publication type; publication date; journal, volume, issue, and page numbers of publication; place of publication (i.e., first author’s institutional address); and digital object identifier. Study design, time frame of study, climate zone of interest, location of study (i.e., country), and region of study (localized when reported) will also be extracted.

Data extracted about participants will include the definition of the target and source populations, size of the target population, and size of the source population. Relevant demographic information (e.g., age, sex) at the study population level will be extracted when reported.

Exposure related data extracted will include the name of the Indigenous population, the size of the Indigenous source population, and the size of the Indigenous study population. For the comparator, the size of the non-Indigenous source population, and the size of the non-Indigenous study population will be extracted. For all studies, the definition provided for Indigenous peoples will be extracted. Further, if causal mechanisms for differences in seasonal or meteorological effects on ARI are provided, these will be extracted.

Information related to the seasonal or meteorological parameter(s) of interest, as well as the ARI outcome(s) of interest will be extracted. In addition, the association(s) between ARI and season or meteorological parameters will be extracted. For season and meteorological parameters, extracted information will include the name of each parameter (e.g., rainfall) and its related measure (e.g., millimeters); type of temporal cycle (e.g., daily, seasonal, annual); and number of cycles completed (e.g., years). Additionally, the source of data used to evaluate the season or meteorological parameter will be collected (e.g., meteorological stations). The extracted information for each ARI outcome will be the specific ARI outcome (i.e., case definition), measurement of the ARI outcome (e.g., self-reported), group-level metric for each population group (e.g., prevalence) and the effect size (i.e., beta) comparing the Indigenous and non-Indigenous peoples or strata (e.g., odds ratio). Where two models are presented (e.g., ARI-exposure models for Indigenous and non-Indigenous peoples), odds ratios will be extracted for each strata (Fig. 1). Where one model is presented (e.g., a model with Indigeneity and weather parameter(s) as fixed effects), the regression coefficients for the season or meteorological parameter and indigeneity will be extracted. In the case that a study presents results for both options (i.e., two strata models and a single model with fixed effects), data will be extracted for both options. For each association, the measure(s) of precision (e.g., standard error of the mean, standard deviation, and/or confidence intervals) will be extracted when provided. If only the p value and sample size are reported, these data will be extracted and a measure of precision will be calculated from the available data for each association.

Finally, information will be extracted on the type of modeling or statistical approach (e.g., linear regression) used, and if and which confounders were considered. Confounders considered will be extracted and a list will be generated for various ARI outcomes. Additionally, the unit of analysis (e.g., individual, household, or community) and spatial resolution of the climate data used for modeling will be extracted.

Process for data extraction

A data extraction form will be created in DistillerSR®. The extraction form will be piloted and tested by the data extractors on a subset of studies (5% of studies if n > 50, 10% of studies if n ≤ 50). Following pilot testing, the form will be adapted as recommended by the extractors to improve usability and completeness. The first author and one additional extractor who each have training in epidemiology and systematic literature review processes will complete data extraction. Data extraction will be completed independently and the extractors will compare the data for consensus. If the extractors cannot answer a question, consensus will confirm that the data are unavailable to answer the question.

In the event that the data presented in a study are unclear, missing, or presented in a non-extractable or unusable form, authors of studies published in the last 5 years (since January 1, 2011) will be contacted for clarification. Authors will be contacted via email, and a follow-up email will be sent 2 weeks later. Authors will be provided 4 weeks from the initial contact to respond. If data from older studies are unclear, missing, or presented in a non-extractable or unusable form, authors will not be contacted. Missing data will be noted in the report.

Risk of bias assessment for eligible studies

The risk of bias (ROB) assessment was adjusted from existing tools (i.e., risk of bias in non-randomized studies of interventions) [25]. In particular, adjustments were needed to account for ecological studies and repeated measures. One question was added to investigate ROB due to ecological studies and one question was added to investigate ROB due to repeated measures. Questions related to experimental interventions were removed. Two reviewers will conduct the ROB assessment independently, in conjunction with data extraction, at the study level (i.e., one ROB analysis will be conducted per study). Both reviewers conducting the ROB assessment will have advanced graduate training in epidemiology and bias assessment. In total, nine domains of bias will be tested according to predetermined criteria for high, low, or unclear ROB. The ROB will be conducted using a form in DistillerSR® with a textbox to record the rationale for selecting the level of ROB for each domain.

Confounders relevant to all or most studies

Since this review will focus on the association between weather parameters and ARI outcomes, confounders in this study are those that affect the association between these weather parameters and ARI in Indigenous versus non-Indigenous peoples. Important confounders that could affect all or most studies are: (i) gender, (ii) age, and (iii) local wealth (e.g., regional or national).

Strategy for data synthesis

The review will analyze information for both associations (systematic review and meta-analyses) and context (e.g., via descriptive statistics, narrative, and descriptive spatial analyses). This approach is necessary to avoid comparing unlike populations, exposures, or outcomes. Analyses will be completed using STATA© version 13.1 and REVMAN© version 5.

Prior to beginning meta-analyses, descriptive statistics will be conducted on extracted data. Frequencies, proportions, and missing data will be considered for each extracted variable. The descriptive data will serve to describe the literature available on this topic and to represent the dataset under study.

Meta-analyses will be conducted for each ARI association (i.e., each ARI outcome and weather parameter identified) that has at least two studies providing data (i.e., two studies presenting the association between the same ARI outcome and same weather parameter). The outcome used for meta-analyses will be the ROR representing the relative effects of weather parameters on ARI between Indigenous and non-Indigenous peoples (Fig. 1). For studies presenting two models (i.e., ARI associations for each strata), odds ratios for each eligible study for each group will be extracted (i.e., Indigenous, non-Indigenous) to calculate a ROR. In studies presenting a single model, regression coefficients will be used directly to calculate odds ratios and solve for the ROR (Fig. 1). The standard error of the ROR will be calculated according to Golder et al. [26] for both study types. Meta-analyses will be conducted using random effects models.

Between-study heterogeneity will be explored using the I 2 statistic (I 2 < 0.25 considered homogeneous). If heterogeneity exists, sources of heterogeneity will be explored by sub-group analyses. For eligible studies, we propose to categorize the studies by (i) population, (ii) outcome, (iii) exposure, and (iv) location. Studies will be categorized by groups of Indigenous peoples (e.g., all studies of Maori peoples) for population; as upper respiratory, lower respiratory, or unclear/both for outcome; as seasonal or meteorological parameters for exposure; and as a high-, middle-, or low-income country and by climate zone for location. When there are at least two effect sizes for each category, we will calculate a summary effect size.

Descriptive statistics will be conducted on each of the domains of the ROB. If enough data are available and ROB profiles vary, heterogeneity will be explored using sub-group analyses on each domain as an independent variable.

Publication bias is the concept that significant results are more likely to be published than non-significant research results [27]. An evaluation of publication bias will be conducted using a funnel plot (if n > 10 studies). Ratio measures of association will be plotted on the logarithmic scale to increase symmetry. A 95% confidence interval will be plotted, and different symbols will be used if heterogeneity is present. Interpretation of the funnel plot will be done visually. A test of significance for publication bias will be conducted using Egger’s test [28].

Finally, to summarize data about region of study and climate zone, descriptive spatial analyses will be performed. Specifically, point maps will be generated. The point map will illustrate specific study locations. When point locations are not provided in text with latitudinal and longitudinal coordinates, study sites will be geo-located using Google Maps© and the best information available in the study. The point map will use open source shape files and will be built in R Spatial© and R Maptools© statistical packages.

Strategy for presentation of the results

The final search strategy for each database and all ancillary searches conducted will be provided in the Additional file 1 of the final report. A flow chart, following the PRISMA guidelines [17], will be used to illustrate where citations were eliminated during screening and ancillary searches, including information about the rationale for exclusion in full article screening.

To illustrate the potential for publication bias and small study effects, a funnel plot will plot the effect estimates (horizontal axes) against the standard error (vertical axes) for each meta-analysis with n > 10 studies [29].

The results of this review will be provided via text and characteristics of studies tables in a published journal article. The tables will describe the seasonal or meteorological exposure, the outcome(s) measured, the direction and magnitude of association, and the period of study. Descriptive statistics (i.e., frequencies, proportions, missing data) will be provided as extensions of this table when appropriate or in narrative (i.e., proportion of studies with each ARI outcome).

Individual study results will be presented in forest plots [17]. If heterogeneity exists, separate forest plots will be used to illustrate results by strata. If the data are too limited (i.e., fewer than two studies with the same population, exposure, outcome, and region) or are heterogeneous, the results will be presented in a forest plot without a summary effect size.

A summary of findings table for key outcomes will be generated based on the Grading of Recommendations Assessment, Development and Evaluation [30]. A priori key outcomes will include prevalence of upper ARI and prevalence of lower ARI. Additional key outcomes identified in the systematic literature review will be documented as protocol amendments.

A table will be provided to summarize the findings of the ROB assessment. This table will follow the ROB presentation suggested in the PRISMA guidelines [17]. An additional column will highlight the rationale for the study’s ROB level.

Maps indicating the specific location of studies (point map) will be generated. Climate zones will be indicated on each map (e.g., tropical, temperate, or arctic).

Ethical considerations

This research does not involve working directly with Indigenous and non-Indigenous communities, but rather with previous research conducted with these communities. In conducting this research, ethical principles will still be at the forefront, and will involve considerations for small population sizes and framing of the findings.


This systematic review protocol presents the method for the synthesis of current evidence related to differences in seasonal or meteorological association with ARI between Indigenous and non-Indigenous peoples living in the same region. This proposed review will likely be the first to summarize the potentially different associations between ARI and weather parameters between Indigenous and non-Indigenous peoples.

The results of the meta-analysis will examine whether Indigenous peoples are equally susceptible to associations between weather parameters and ARI, and whether this relationship varies by place, population, or other characteristics. A deeper understanding of this relationship will advance the academic literature and potentially lead to intervention strategies as climate change progresses. Further, an understanding of the differences between Indigenous and non-Indigenous communities can aid in planning, resource allocation, and determination of appropriate interventions.



Acute respiratory infections


International Work Group for Indigenous Affairs


Population, exposure, comparator, and outcome


Preferred Reporting Items for Systematic Review and Meta-Analysis


Risk of bias


Ratio of odds ratios


United Nations Refugee Agency


  1. Mulholland K. Global burden of acute respiratory infections in children: implications for interventions. Pediatr Pulmonol. 2003;36(6):469–74.

    Article  PubMed  Google Scholar 

  2. Nair H, Nokes DJ, Gessner BD, Dherani M, Madhi SA, Singleton RJ, O’Brien KL, Roca A, Wright PF, Bruce N. Global burden of acute lower respiratory infections due to respiratory syncytial virus in young children: a systematic review and meta-analysis. Lancet. 2010;375(9725):1545–55.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Nair H, Simões EA, Rudan I, Gessner BD, Azziz-Baumgartner E, Zhang JSF, Feikin DR, Mackenzie GA, Moiïsi JC, Roca A. Global and regional burden of hospital admissions for severe acute lower respiratory infections in young children in 2010: a systematic analysis. Lancet. 2013;381(9875):1380–90.

    Article  PubMed  PubMed Central  Google Scholar 

  4. Hall KKC, Anne B. Sloots, Theo P. Anderson, Jennie Kemp, AnitaHammill, JanOtim, Michael, O’Grady K-AF. The respiratory health of urban Indigenous children aged less than 5 years: study protocol for a prospective cohort study. BMC Pediatr. 2015, 15(56):8.

  5. Murdoch KM, Mitra B, Lambert S, Erbas B. What is the seasonal distribution of community acquired pneumonia over time? A systematic review. Australasian Emerg Nurs J. 2014;17(1):30–42.

    Article  Google Scholar 

  6. du Prel J-B, Puppe W, Gröndahl B, Knuf M, Weigl F, Schaaff F, Schmitt H-J. Are meteorological parameters associated with acute respiratory tract infections? Clin Infect Dis. 2009;49(6):861–8.

    Article  PubMed  Google Scholar 

  7. Vlahov D, Freudenberg N, Proietti F, Ompad D, Quinn A, Nandi V, Galea S. Urban as a determinant of health. J Urban Health. 2007;84(1):16–26.

    Article  PubMed Central  Google Scholar 

  8. Cohen S. Social status and susceptibility to respiratory infections. Ann N Y Acad Sci. 1999;896(1):246–53.

    Article  CAS  PubMed  Google Scholar 

  9. Ford JD. Indigenous health and climate change. Am J Public Health. 2012;102(7):1260–6.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Ye Y, Zulu E, Mutisya M, Orindi B, Emina J, Kyobutungi C. Seasonal pattern of pneumonia mortality among under-five children in Nairobi’s informal settlements. Am J Trop Med Hyg. 2009;81(5):770–5.

    Article  PubMed  Google Scholar 

  11. Willox AC, Harper SL, Edge VL, Landman K, Houle K, Ford JD. The land enriches the soul: on climatic and environmental change, affect, and emotional health and well-being in Rigolet, Nunatsiavut, Canada. Emotion Space Soc. 2013;6:14–24.

    Article  Google Scholar 

  12. Willox AC, Harper SL, Ford JD, Landman K, Houle K, Edge VL. “From this place and of this place:” climate change, sense of place, and health in Nunatsiavut, Canada. Social Sci Med. 2012;75(3):538–47.

    Article  Google Scholar 

  13. United Nations: Indigenous peoples, Indigenous voices factsheet. Edited by Issues UPFoI. New York City: United Nations; n.d.: 2.

  14. Berrang-Ford L, Dingle K, Ford JD, Lee C, Lwasa S, Namanya DB, Henderson J, Llanos A, Carcamo C, Edge V. Vulnerability of Indigenous health to climate change: a case study of Uganda’s Batwa Pygmies. Soc Sci Med. 2012;75(6):1067–77.

    Article  PubMed  Google Scholar 

  15. Ohenjo N, Willis R, Jackson D, Nettleton C, Good K, Mugarura B. Health of Indigenous people in Africa. Lancet. 2006;367(9526):1937–46.

    Article  PubMed  Google Scholar 

  16. Marmot M. Social determinants of health inequalities. Lancet. 2005;365(9464):1099–104.

    Article  PubMed  Google Scholar 

  17. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264–9.

    Article  PubMed  Google Scholar 

  18. Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P, Stewart LA. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Rev. 2015;4(1):1.

    Article  Google Scholar 

  19. Naumova E, Jagai J, Matyas B, DeMaria A, MacNeill I, Griffiths J. Seasonality in six enterically transmitted diseases and ambient temperature. Epidemiol Infect. 2007;135(02):281–92.

    Article  CAS  PubMed  Google Scholar 

  20. Willett HC, Sanders F. Descriptive meteorology: New York: Elsevier; 2013.

  21. Bartlett JG, Madariaga-Vignudo L, O’Neil JD, Kuhnlein HV: Identifying Indigenous peoples for health research in a global context: a review of perspectives and challenges. Int J Circumpolar Health. 2007:66(4):287-307.

  22. Dasaraju PV, Liu C. Infections of the respiratory system. Medical Microbiology 4th ed. Galveston: University of Texas; 1996.

  23. Hedgeman E, Lipworth L, Lowe K, Saran R, Do T, Fryzek J. International burden of chronic kidney disease and secondary hyperparathyroidism: a systematic review of the literature and available data. Int J Nephrol. 2015;2015:184321.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Gehanno J-F, Rollin L, Le Jean T, Louvel A, Darmoni S, Shaw W. Precision and recall of search strategies for identifying studies on return-to-work in Medline. J Occup Rehabil. 2009;19(3):223–30.

    Article  PubMed  Google Scholar 

  25. Sterne JA, Hernán MA, Reeves BC, Savović J, Berkman ND, Viswanathan M, Henry D, Altman DG, Ansari MT, Boutron I, et al. The Risk Of Bias In Non-randomized Studies – of Interventions (ROBINS-I) assessment tool. 2016.

    Google Scholar 

  26. Golder S, Loke YK, Bland M. Meta-analyses of adverse effects data derived from randomised controlled trials as compared to observational studies: methodological overview. PLoS Med. 2011;8(5):e1001026.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Duval S, Tweedie R. Trim and fill: a simple funnel-plot–based method of testing and adjusting for publication bias in meta‐analysis. Biometrics. 2000;56(2):455–63.

    Article  CAS  PubMed  Google Scholar 

  28. Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ. 1997;315(7109):629–34.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, Savović J, Schulz KF, Weeks L, Sterne JA. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Oxman AD, Group GW. Grading quality of evidence and strength of recommendations. BMJ. 2004;328(19):1490–4.

    PubMed  Google Scholar 

Download references


The authors wish to thank Ali Versluis, University of Guelph Library, for her time and expertise contributed to this protocol.


Funding for this protocol was provided through scholarships from the Ontario Veterinary College (K. Bishop-Williams) and the Ontario Graduate Scholarship program (K. Bishop-Williams).

Availability of data and materials

Not applicable.

Authors’ contributions

KBW will serve as the first author of the protocol and review paper. She led all stages of protocol development, including development of the research question and objectives, search strategy, and extraction and analysis plans. JMS, LBF, VLE, AC, and SLH supervised and contributed to the entirety of the development of plans for searching, screening, extracting, and writing phases. All authors have graduate training in epidemiology and/or health studies. All authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Not applicable.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Katherine E. Bishop-Williams.

Additional file

Additional file 1:

The Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) 2015 checklist recommended items to address in a systematic review protocol with red text to demonstrate where information can be found in the body text. (DOCX 19 kb)

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bishop-Williams, K.E., Sargeant, J.M., Berrang-Ford, L. et al. A protocol for a systematic literature review: comparing the impact of seasonal and meteorological parameters on acute respiratory infections in Indigenous and non-Indigenous peoples. Syst Rev 6, 19 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: