The characteristics of behaviour change interventions used among Pacific people: a systematic search and narrative synthesis - BMC Public Health
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Matenga-Ikihele et al. BMC Public Health (2021) 21:435
https://doi.org/10.1186/s12889-021-10420-9
RESEARCH ARTICLE Open Access
The characteristics of behaviour change
interventions used among Pacific people: a
systematic search and narrative synthesis
Amio Matenga-Ikihele1,2* , Judith McCool1, Rosie Dobson3, Fuafiva Fa’alau4 and Robyn Whittaker3
Abstract
Background: Pacific people living in New Zealand, Australia, United States, and the Pacific region continue to
experience a disproportionately high burden of long-term conditions, making culturally contextualised behaviour
change interventions a priority. The primary aim of this study was to describe the characteristics of behaviour
change interventions designed to improve health and effect health behaviour change among Pacific people.
Methods: Electronic searches were carried out on OVID Medline, PsycINFO, PubMed, Embase and SCOPUS
databases (initial search January 2019 and updated in January 2020) for studies describing an intervention designed
to change health behaviour(s) among Pacific people. Titles and abstracts of 5699 papers were screened; 201 papers
were then independently assessed. A review of full text was carried out by three of the authors resulting in 208
being included in the final review. Twenty-seven studies were included, published in six countries between 1996
and 2020.
Results: Important characteristics in the interventions included meaningful partnerships with Pacific communities
using community-based participatory research and ensuring interventions were culturally anchored and centred on
collectivism using family or social support. Most interventions used social cognitive theory, followed by popular
behaviour change techniques instruction on how to perform a behaviour and social support (unspecified).
Negotiating the spaces between Eurocentric behaviour change constructs and Pacific worldviews was simplified
using Pacific facilitators and talanoa. This relational approach provided an essential link between academia and
Pacific communities.
Conclusions: This systematic search and narrative synthesis provides new and important insights into potential
elements and components when designing behaviour change interventions for Pacific people. The paucity of
literature available outside of the United States highlights further research is required to reflect Pacific communities
living in New Zealand, Australia, and the Pacific region. Future research needs to invest in building research
capacity within Pacific communities, centering self-determining research agendas and findings to be led and
owned by Pacific communities.
Keywords: Pacific health; health behaviour change, systematic review, talanoa
* Correspondence: a.ikihele@auckland.ac.nz
1
Epidemiology and Biostatistics, School of Population Health, University of
Auckland, Private Bag 92019, Auckland 1142, New Zealand
2
Moana Research, Nga Hau Māngere Birthing Centre, 14 Waddon Place,
Auckland, Māngere, New Zealand
Full list of author information is available at the end of the article
© The Author(s). 2021 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this article are included in the article's Creative Commons
licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons
licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the
data made available in this article, unless otherwise stated in a credit line to the data.Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 2 of 15 Background ‘Pacific people’ is a broad, collective term used to Pacific people experience a disproportionately high describe the dynamic and diverse groups of people from burden of long-term conditions and the associated risk the sub-regions of Oceania: Polynesia, Melanesia, and factors [1–3], despite being minority communities in Micronesia. Distinct from a Western appreciation of major countries such as the United States (USA), health, health and wellbeing for Pacific people is holistic Australia, and New Zealand. Cardiovascular diseases, and intrinsically related to identity, land, cultural values, diabetes, cancer, and respiratory diseases represent a roles and responsibilities [15–17]. Although there are leading threat to human health and are the largest cause similar values observed within each Pacific Island, such of premature mortality in the Pacific [4]. The burden of as reciprocity, respect, relationships and collectivism, multi-morbidity is also more significant, with people there are vast linguistic, cultural, geographical, migration experiencing two or more conditions requiring complex and political differences between different Pacific Islands care and management [5]. These ongoing health inequities [16–18]. This diversity is further enhanced with place of have persisted for several decades [6, 7] irrespective of large birth (island born vs diaspora), and intermarriage where investments to improve Pacific health outcomes [3, 5]. Pacific people now identify with two or more ethnic There is a growing appreciation of how adopting life- groups [14, 16, 17]. style changes, such as being smoke-free, a healthy diet, The family unit is an important institution for Pacific regular physical activity, and limiting alcohol consump- people, which extends beyond the nuclear family and is tion, can positively affect health outcomes [8]. Even inclusive of village and church groups [19–22]. Aligning small changes in these health behaviours are known to with the holistic Pacific Fonofale model [23], family re- have overall positive health gains [9]. The challenges mains an important foundation for Pacific people, pro- when undertaking initial health behaviour changes and viding a stable system of support despite social and eventually maintenance recognises that identifying key economic changes, transnational migration, urbanisation characteristics that influence positive health change and modernisation [15, 24]. An individual’s motivation among Pacific people is necessary. to change behaviour is strongly interwoven with the Behaviour change interventions often consist of mul- socio-cultural roles and family responsibilities [20, 25– tiple components designed to effect change in both an 27]. Several studies have highlighted Pacific people to individual’s perceptions (cognitions) and behaviours perceive health as being able to provide for their families [10]. These interventions are often underpinned by the- and the wellbeing of their family unit, rather than the oretical frameworks and may incorporate Behaviour physical ailments of an individual [21, 28, 29]. Moreover, change techniques (BCTs); methods identified as the ‘ac- studies that have included family and social support have tive ingredients’ of an intervention (e.g. goal setting) shown to be more effective than those requiring individ- [11]. Behaviour change interventions that are grounded uals to undertake behaviour change on their own accord in theory are generally accepted to be more effective [30]. Understanding the determinants of behaviour than those that lack a theoretical basis [10–12]. change for Pacific people is a necessary step towards de- Health behaviour theories provide a framework for in- signing interventions that are culturally relevant, salient terventions designed to alter health-related cognitions and sustainable [14]. and behaviours [12, 13]. Most theoretical frameworks Acknowledging the cultural differences, values, beliefs, used in public health and behaviour change interven- structures, practices and worldviews of health and well- tions are predicated on the individual changing their be- being unique to Pacific people can provide the context- haviours [10]. These models, developed within ual framework for developing health behaviour Eurocentric perceptions of human behaviour, are built interventions [25, 28]. Fundamental to the success of on the assumption that behaviour is ultimately individ- any programme centres around creating interventions ual and rational. This position dismisses other drivers to that are acceptable and accessible for Pacific people, behaviour – namely, the spiritual, social, cultural, and who have “the right to the highest attainable standard of environmental (collective) variables [12–14]. Social and health” in New Zealand [31]. Evidence suggests that cultural dynamics are also often sidelined within main- traditional or non-Pacific programmes have not been stream health behaviour change models. Understanding effective, despite considerable investments [22, 27, 32]. health behaviours and the context in which these occur Furthermore, it is important to consider the role of inad- is vital to designing effective interventions for minority equate workforce capability in cultural diversity, institu- and indigenous populations. Contemporary models of tional racism, and unconscious bias have contributed to the health behavior change are needed, which are inclusive state of Pacific health in New Zealand [3, 15, 22, 27, 33]. inclusive, and reflect different culturally ascribed values, Pacific led programmes reflect the importance of rela- including wider spiritual, socio-cultural, and environ- tional rather than individualistic relationships among mental influences [13] for Pacific people. Pacific people. The significance of relationships can be
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 3 of 15
understood using the Samoan concept of ‘teu le va’, Eligibility criteria
which is premised heavily on the relational contexts be- Studies were eligible to be included if they described an
tween people, things and the environment, as well as the intervention designed to change health behaviour(s) in
nurturing and protection of mutually respectful relation- Pacific people. Included studies were required to report
ships over individual agendas [15, 34, 35]. The culturally a clinical or behaviour change outcome with or without
located concept of ‘va’ (relationships) cannot be mea- randomised controlled conditions. Studies were also re-
sured and is not inherently visible within western frame- quired to have at least 40% of participants identifying
works. When this ‘space’ is appropriately nurtured, with a Pacific ethnicity and include participants aged
respect and trust ensue, translating into a higher ac- over 16 years. Studies were excluded if they were not
ceptance of programme initiatives and health care published in the English language or the full text was
[15]. It is also acknowledged in practical terms, such not available. There was no time period restriction
as when health workers or providers are from the applied to the search.
same ethnicity or cultural background, interactions
are reciprocated with respect, and better health out-
comes are achieved [14, 22, 27]. Information sources
Community inspired behaviour change requires the The OVID Medline, PsycINFO, PubMed, Embase and
adoption of Community Based Participatory Research SCOPUS databases were searched during January and
(CBPR) methods, a collaborative approach used in February 2019, and again in January 2020 by the first
public health with minority communities [10, 36]. author (AI). Additional articles were identified from
True to the CBPR methodology, community mem- reference lists. A manual search of the New Zealand
bers, organisations, and researchers become equal Medical Journal, Pacific Health Dialog, Pacific Health
partners in all aspects of the research process [36], Review, Fiji Journal of Public Health and the Australian
addressing health from a holistic perspective. It and New Zealand Journal of Public Health was also
builds trust among those who may have had negative conducted during March 2019 and January 2020 for
experiences, and distrust researchers and the re- further articles that may not have been included during
search process [10]. CBPR also aligns with several the initial database searches.
core Pacific principles: reciprocity, nurturing rela-
tionships, respect, collectivism, and communitarian- Search strategy
ism [16, 17]. Valuing the partnerships between The search strategy used in Medline is provided in Table 1.
academia, health professionals and communities The keywords, their synonyms and various spellings were
through CBPR has the potential to design and facili- reviewed and agreed upon by three co-authors prior to
tate health behaviour change within the socio- commencement. Terms used included 1) Pacific population
cultural context of communities [10]. Current Pacific groups and phrases related to 2) health behaviour change.
health research guidelines [16] emphasise the need Search terms were entered according to each database’s
for such approaches, which aligns with indigenous requirements and using the Boolean operator of and/or to
guidelines developed for Māori in New Zealand [37], combine terms. Database specific filters such as human
Aboriginal and Torres Strait Islanders in Australia population, English language, and full-text articles were
[38] and First Nations communities in Canada [39]. applied where available.
Improving the health and wellbeing for Pacific people
is an important equity issue [3, 15, 28]. Engaging mean-
ingfully and negotiating the spaces between Eurocentric Study selection
theoretical frameworks and Pacific cultural constructs of All articles were downloaded into RefWorks ProQuest
health and wellbeing [40] is needed if evidence-based software, and duplicates were removed. Articles were
and culturally contextualised behaviour change interven- screened for eligibility based on their title and abstracts
tions are to be adopted and sustained. The aim of this by the first author (AI) and sorted accordingly. Full-text
systematic search and narrative synthesis is to describe articles were then retrieved for the remaining papers and
the behaviour change components used in interventions non-relevant studies excluded based on the inclusion
designed to improve health and effect health behaviour criteria. The quality of studies were appraised using the
change among Pacific people. Joanna Briggs Institute (JBI) Critical Appraisal Checklist
[42], a quality assessment tool to determine the meth-
Methods odological quality of quantitative studies. Studies were
The systematic review was informed by the Preferred appraised by the first author (AI) and crosschecked by
Reporting Items for Systematic Reviews and Meta- two co-authors (JM and RD) to determine the final deci-
Analyses (PRISMA) reporting guidelines [41]. sion, and a consensus on inclusion or exclusion reached.Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 4 of 15 Table 1 OVID Medline search strategy
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 5 of 15
Data extraction residing in the state of Hawai’i. Nine studies [43, 44, 46–
Using the PRISMA checklist [41], study information 48, 50–52, 64] used a multiethnic cohort comparing Pa-
including study characteristics, participant information, cific (Native Hawaiians, Samoan, Tongan, Chuukese),
outcomes and behaviour change theoretical frameworks Asian, Caucasian or other non-Pacific population
were extracted and included in a Microsoft Excel groups. Seven studies focused on Pacific only cohorts,
spreadsheet. Study characteristics included authors, year such as Native Hawaiian and other Pacific Islanders [45,
of publication, targeted health condition, the country in 53, 55, 56], Tongan and Samoan [62] and, Tongan, Niue
which the study was conducted, and sample sizes. Par- and Cook Island [61] and Pacific Islanders (mixed) [67].
ticipant information included age, gender, and ethnicity. Two studies compared Pacific and Indigenous Māori
For each study BCTs were identified and coded using [65, 66]. Nine studies were ethnic-specific focusing
the behaviour change taxonomy [11]. The 93 BCTs were only on Native Hawaiians [49, 51], American Samoan
rated as either present (1) or absent (0). BCTs were [58, 59], Marshallese [57, 60], and Samoan [63, 68,
coded when there was clear evidence of inclusion despite 69] populations. Studies were published between 1996
the wide-ranging terminology used. For instance, when and 2020. Table 2 shows further details of the in-
the interventions mentioned education around “discuss- cluded studies.
ing complications” or “reducing risk”, it was coded as Among the twenty-seven studies included in the
information on health consequences (5.1). Social support review, eleven were randomised controlled trials [44, 45,
(general) (3.1) was also coded as such, even when it was 48, 53, 55, 58, 59, 65, 66, 69] and thirteen were non ran-
not always explicitly mentioned. For example, if studies domised studies [43, 46, 47, 49, 51, 52, 56, 57, 60, 62, 64,
required family members or a significant other to be 67, 68]. Other studies used a quasi-experimental design
part of the intervention, or the studies that delivered [63], a prospective uncontrolled cohort study [61] and a
group sessions emphasised group support as an im- descriptive intervention study [50]. The sample sizes
portant component for behaviour change. When ranged from 10 to 675 participants, with an age range
BCTs were not explicitly described as part of the between 13 and 102 years. Across all twenty-seven stud-
intervention, codes were obtained from curriculum ies, the samples included more than 60% female
lesson outlines where available. participants.
Interventions focused mainly on a single health con-
Results dition or specific behaviour. Ten studies increased
A total of 5941 records were identified using the awareness and targeted health behaviour change in
search strategy described. An additional 30 articles people with diabetes [45, 48, 50, 57–60, 62, 64, 68]
were identified by searching references. After remov- and eight studies targeted weight loss [43, 44, 46, 47,
ing duplicates, 5699 articles were reviewed by title 51, 54, 63, 67]. Others targeted chronic disease self-
and abstract of which 5461 were excluded. The full management [52], and health-related behaviour [66],
text of 208 articles were reviewed with articles ex- hypertension [53, 65], smoking cessation [69], heart
cluded if they were not an intervention study (n = failure [55], renal disease [61], physical activity [56],
132), studies focusing on participants aged under 16 and physical activity and nutrition combined [49].
years (n = 39), less than 40% of participants identi- Overall, diabetes was the most common long term
fied as Pacific (n = 27) and full-text articles not condition focused in the studies.
available (n = 10). Twenty-seven articles met the in-
clusion criteria and are included in this review. Fig- Theoretical frameworks
ure 1 shows the study selection process. Five of the Fifteen studies stated they used a theoretical behaviour
included studies reported on the same parent inter- change framework in their intervention. Social cognitive
vention - the Pili Ohana Project (POP). These stud- theory (SCT) was most frequently used (nine studies)
ies are included separately as they each provide [43–47, 51, 53, 54, 67], with one study combining a
unique data and use different cohort samples and range of theories such as participatory metatheory, the
intervention settings. health belief model, SCT and the Pacific talanoa ap-
proach [67]. One study combined the theoretical do-
Study characteristics mains framework with the Fonofale (Pacific) and Te
Fifteen studies were conducted in the USA - Hawai’i Whare Tapa Whā (Māori) health models [66]. Other
[43–55], California [56] and Arkansas [57], as well as its studies were informed by the trans-theoretical model
associated territories of American Samoa [58, 59] and [49], social support model [48, 57], Precede-Proceed
the Marshall Islands [60]. Seven were based in New model [58], and self-efficacy framework [52]. Various
Zealand [61–67] including one in Australia [68] and one combinations of BCTs were incorporated across all 27
in Samoa [69]. Most studies involved Native Hawaiians studies.Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 6 of 15 Fig. 1 PRISMA study selection process Behaviour change techniques solving / coping planning (n = 14), self-monitoring of Using the behaviour change taxonomy [11], 27 different outcome(s) of behaviour (n = 13), modelling of the be- BCTs (from a total of 93) were identified and coded haviour (n = 13), and goal setting (behaviour) (n = 11). across all the interventions. Each study incorporated Each study reported changes in clinical and behavioural BCTs using different combinations, ranging from three outcomes with the various BCTs used. Due to hetero- to 18. The most common BCT was instruction on how geneity of outcome measures, a meta-analysis was not to perform a behaviour (n = 27), followed by social sup- undertaken to determine the effectiveness of BCTs used port (general) (n = 25) and behaviour rehearsal/practice in each study. (n = 21) as seen in Table 3. The two studies that did not provide social support instead focused on behaviour Intervention elements change solely on the individual, despite other interven- All but one study [65] culturally adapted or added cul- tions delivered to individuals incorporating social sup- tural elements to the design of their programme which port from family and friends. Other commonly used closely resonated with Pacific people. Examples included BCTs included health consequences (n = 16), problem- cooking with local or culturally appropriate food, using
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 7 of 15 Table 2 Characteristics of included studies
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 8 of 15 Table 2 Characteristics of included studies (Continued)
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 9 of 15 hula (cultural dance) as a form of exercise, incorporating always worked under the supervision of the research or local language, and cultural customs during education project lead, and alongside a multidisciplinary team sessions, and using a minister or kupuna (elder) to lead (nurses, physicians, nutritionists and pharmacists). A and close education sessions. One study used co-design multidisciplinary approach was used in eight studies [66] with Pacific (including Māori) communities to [43, 45–48, 51, 53, 65] where a CHW, PHE or HCA design the behaviour change intervention. Another study facilitated sessions or worked alongside health pro- was informed by a Pacific nutrition train the trainer fessionals (nurse specialist, physician, pharmacist, workshop for Pacific people living in New Zealand [67]. nutritionist, physiotherapist) including a kumu hula Thirteen studies were adapted for Native Hawaiian’s and (hula expert). Three studies [58, 59, 65] used a Pacific Islanders living in Hawai’i [43–55], four were treatment protocol which determined the length and adapted for Samoans residing in American Samoa [58, 59], frequency of visits by a CHW, PHE or HCA. When Australia [68] and Samoa [69], and two studies were participants clinical observations were elevated (e.g. adapted for Marshallese Islanders living in Arkansas [57] high blood pressure), the CHW, PHE or HCA and the Marshall Islands [60]. Ten studies [43, 44, 46, 47, would communicate their results with their phys- 53, 54, 56, 57, 67, 68] stated they used a CBPR approach ician for further clinical examination or treatment where research leads partnered with Pacific communities adjustments. and organisations who informed the cultural and commu- Eighteen studies delivered their intervention using nity adaptation of programmes. While most interventions group-based models [43–48, 51–57, 60, 62–64, 67, 68] were delivered in English, seven studies offered a bilingual and nine were delivered to individuals [49, 50, 55, 58, 59, approach: one study offered English, Chuukese and Samoan 61, 65, 66, 69]. The intensity, duration, and time [44], one offered English, Tongan and Samoan [62] and five between intervention sessions varied. One study ran for offered English and Samoan [58, 59, 63, 68, 69]. To ensure a week [69], eight studies lasted twelve weeks [43, 45– fidelity of the programme once culturally adapted, only 48, 51, 53, 66], with others lasting six-eight weeks [52, three studies mentioned research leads observed their facili- 56, 57, 67], four months [60], six months [44, 64, 68], tators [45, 50, 52]. nine months [54], twelve months [49, 50, 55, 58, 59, 63, Thirteen studies delivered their interventions alongside 65] and two years [61, 62]. Each session lasted between community-based organisations [43–46, 48, 49, 51–53, one to two and a half hours. Intervention follow-up 56, 59–61], three were based in churches [62, 63, 68], ranged from three to 24 months. Six studies did not de- three were delivered in participants workplaces [47, 54, fine how many hours each visit or education session 64] and one in a community location [67]. Two studies lasted [47, 49, 50, 58, 59, 68]. Three interventions used visited participants in their homes [55, 65] and two used multimedia with one utilising a diabetes educational mobile phones with one using an app [66] and the other video to compliment the physical activity sessions deliv- delivering short message service (SMS) behaviour ered [64], one used a DVD and workbook for partici- change messages to participants [69]. Three studies en- pants [54], while another intervention used video as the abled participants to choose the location of their inter- primary mechanism lasting 10 min [56]. Despite the lat- vention, selecting either a church, community-based ter study lasting two-months, and targeting church organisation, workplace or home [50, 57, 58]. Only eight groups and community organisations, it was unclear studies encouraged participants to include family mem- how many sessions participants viewed the video. bers or a significant other to be involved with education sessions [44, 48, 50, 55, 57, 58, 66, 68]. Outcomes Eleven studies used peer health educators (PHE) [43– Twenty-six studies reported positive outcomes and im- 47, 49, 51, 54, 60, 62, 63] to facilitate education sessions, provements in intended behaviour change measures. four studies used community health workers (CHW) One study however found no significant difference be- [50, 53, 57, 59], one used a health care assistant (HCA) tween the intervention and control groups to health- [65] while another study used the term community related behaviours (physical activity, smoking behaviour, coach facilitators (CCF) [68]. Of the three studies based alcohol intake, fruit and vegetable intake) [66]. Five stud- in churches, one used PHE [68] to deliver sessions to ies reported improvements in nutritional intake [46, 47, church members, while two [62, 63] set up health com- 49, 60, 67] and eight studies reported weight loss mittees led by ministers wives and members within the changes [43, 44, 46, 47, 51, 54, 60, 67]. Improvements in church. These health committees then selected church physical activity were reported in nine studies [46, 47, members to become trained CHW. CHW and PHE were 49, 52, 56, 62–64, 68]. In one study [64], physical activity either trained by their research or project leads or in the intervention group improved significantly com- undertook formal education through a tertiary institu- pared to the control group while in another study [63], tion [50, 62–64, 68]. The PHE, CHW, HCA and CCF physical activity rates only improved in some
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 10 of 15 Table 3 Frequency of behaviour change techniques used participants, whereas those who reported they were sed- Reported changes in Haemoglobin A1c (HbA1c) var- entary at baseline remained unchanged post- ied from baseline to follow up in six studies [45, 48, 50, intervention. One study [62] which ran two intervention 57, 60, 68]. One study [50] reported a more significant and control groups (one Samoan and one Tongan reduction in HbA1c for the intervention group than the church) reported weight, waist circumference and exer- control group, and three studies [45, 57, 60] reported cise improvements in only one of the intervention positive changes post-intervention. Another study [48] groups (Samoan), and not the second intervention group however reported participant’s glycaemic control at six (Tongan) or either control groups. months was not significantly different from those in the
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 11 of 15
control group despite initial improvements after the first participants to share and exchange their knowledge,
three months. emotions and experiences. Talanoa (conversation, a talk,
Nine studies reported improvements in blood pres- an exchange of ideas or thinking) provides a culturally
sure [46–48, 51, 53, 58, 60, 61, 65] and six studies appropriate approach for ‘va’ (relationships) to be estab-
reported improvements in self-management and dia- lished and nurtured between researchers, facilitators,
betes knowledge [45, 48, 60, 62, 64, 68]. Other reported and participants. The nurturing of these relationships
improvements included self-reported health [52], self- creates a space where talanoa or social conversations
efficacy [47, 52], smoking cessation [69], medication ad- can take place, holistically intermingling the knowledge,
herence [60], improved cholesterol levels [51], and in- experiences and emotions, shared between researchers
creased primary care physician visits compared to and participants [72]. Talanoa constitutes a culturally
emergency department visits [59]. Satisfaction and appropriate method which Pacific researchers have pri-
acceptability with the interventions were reported in marily used to engage with Pacific communities [34, 16].
four studies [52, 56, 62, 64]. One study [58] used co-design with Pacific and Māori
communities, complimenting CBPR. Co-design em-
Discussion powers users to tailor interventions according to their
The aim of this review was to describe behaviour change cultural needs and context from design inception [64].
components used in interventions to improve health and As such, participants in this study aligned their well-
effect health behaviour change among Pacific people. To being priorities with ethnic-specific models of health and
our knowledge, it is the first study to highlight and de- wellbeing, namely Fonofale [23] (Pacific) and Te Whare
scribe the theoretical underpinnings and BCTs used in Tapa Whā (Māori) [73]. Despite co-designing a cultur-
interventions designed to improve health among Pacific ally tailored, lifestyle support intervention, the authors
people. Twenty-seven studies met the inclusion criteria noted participation in the control and intervention
for the review. Most studies focused on diabetes and groups did not affect adherence to health-related behav-
weight loss, followed by hypertension, physical activity, iour guidelines. Perhaps considerations regarding digital
and smoking cessation. inclusion and the digital health literacy skills required
An important feature to highlight was the collabora- for this intervention were overlooked, which are known
tive CBPR approach used to culturally adapt interven- equity issues for Pacific people in New Zealand [74, 75].
tions. When CBPR was not used, studies partnered All studies in this review incorporated cultural adapta-
with church-based organisations, workplaces, or local tions and elements, and utilised BCTs of some sort.
communities, evident from 1996 until 2020. An elem- Using Michie’s behaviour change taxonomy [11], 27
ent that could strengthen partnerships and provide BCTs (out of a possible 93) were identified and used in
self-determination with future research agendas is the different combinations across all studies. The minimum
need to continue building research capacity and cap- number of BCTs used in a study was three, the majority
ability among Pacific communities. This ensures being 18 techniques. Specific behaviours targeted in-
future research builds on existing capacity within cluded combinations of physical activity, healthy eating,
Pacific communities rather than duplicating efforts self-management, medication adherence, problem-
with external research agencies. To do this, meaning- solving, coping, and increasing knowledge of health con-
ful partnerships with Pacific communities using CBPR ditions. The most popular BCT was instruction on how
must be established, which demonstrates cultural in- to perform a behaviour (4.1), used to demonstrate cul-
tegrity, rigour and acknowledgement of Pacific world- turally appropriate meals, facilitate exercise classes and
views and values. Moreover, Pacific communities self- provide educational sessions around the targeted health
determination for Pacific research is necessary to conditions and behaviours; followed by social support
improve health equity [14, 36] and culturally safe (unspecified) (3.1). Providing social support resonates
research practices [16, 70, 71]. with Pacific values, drawing strength from socio-cultural
Despite evidence of the impact on behaviour change, relationships within their collective contexts, such as
most interventions were short-term with varying study extended family, community, and church networks
designs and little regard to sustainability. Only two stud- [17, 22, 28].
ies followed up on weight maintenance after completing Most studies in this review were based in the USA or
a three-month weight loss programme [44, 54]. Most in- its affiliate countries (American Samoa and the Marshall
terventions were also centred on Eurocentric theoretical Islands). Only seven were from New Zealand and one
frameworks, namely SCT. While studies are considered from Australia and Samoa. All but two of the studies
more effective when incorporating such theoretical com- were representative of Pacific countries from Polynesia,
ponents [10–12], only one study [67] included the with another from Micronesia (Marshall Islands).
Pacific talanoa approach as a means of allowing Importantly, there were no studies from MelanesianMatenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 12 of 15 countries that met our search criteria, despite evidence Although most interventions were delivered alongside showing they too experience high rates of long-term community-based organisations, alternative sites such as conditions [76]. Pacific people throughout the diaspora workplaces, churches, homes, and the use of multimedia are diverse with different contexts and cultural and mobile phones offer non-traditional approaches that constructs, language, migration histories, constitutional are also acceptable for Pacific communities. Incorporat- ties and health needs [16, 17]. Even though Pacific ing traditional cultural art forms such as hula also draw people share many commonalities, they are not a on the cultural nuances that have resonated with Pacific homogenous group. communities for centuries. The wide age range of 13 A key component for more than half the studies, and 102 years was only evident in the group delivered which needs to be acknowledged, was the utilisation of interventions based in churches. Though not the focus Pacific health workers (CHW, PHE, HCA) from partici- of this review, including young people in interventions pants own communities. Even if they were not leading alongside their parents and grandparents, enables an intervention components, they worked within a multidis- intergenerational approach to improving health and ciplinary team who provided supervision and support. In wellbeing within Pacific families [77]. Furthermore, evi- one study [53], PHE worked alongside a kumu hula dence suggests that exercise among older people is pro- (hula expert), who delivered hula lessons while PHE fo- tective against fall-related fractures [78]. More studies cused on the education modules. Another study [62] reporting on the long-term effects of BCTs would be found church members were more connected to the fa- beneficial to determine factors supporting sustained cilitators from their church than those who were not, behaviour change among Pacific populations. which is essential to consider. Few differences were found between the different roles (CHW, PHE, HCA) as Limitations all were required to undertake training before working A key limitation of this review is due to heterogeneity of with participants. Studies that used Pacific health participants, types of interventions and outcomes, a workers provided an important link between academia meta-analysis was not completed. Therefore, conclusions and the community, assisting with the cultural adapta- cannot be drawn on the effectiveness of behaviour tion of programmes and supporting participants through change interventions among Pacific people. All eligible behaviour change. Bilingual programmes may be more studies focused on Pacific populations from Polynesia effective by overcoming language barriers, especially living in the USA, American Samoa, and New Zealand. when interventions are delivered predominately in Eng- One study focused on the Marshall Islands, Australia, lish. Pacific health workers’ added value was their ability and Samoa, which limits the generalisability of findings. to connect all the elements within each study (i.e. lan- This review was also limited to studies with more than guage, ethnicity, BCT, facilitating education sessions) ef- 40% of the study population identified as Pacific. This fectively enhancing social support and promoting would have excluded other studies that included a positive behaviour change outcomes. Building capacity smaller sample of Pacific people. among Pacific communities to self-determine their re- search aspirations is essential for community-led and Conclusion owned research [34, 70, 71]. This review provides new and important insights into A positive outcome, not well represented in the litera- potential elements and components when designing be- ture, was the minister’s wives role as facilitators within a haviour change interventions for Pacific people. It also church setting. Minister’s wives were key to establishing highlights the paucity of literature available for Pacific health committees and facilitating nutrition and exercise communities living outside of the USA. programmes. Policy change was also possible within the Future behaviour change research with Pacific com- church context. One church [62] created policies to in- munities should be community created and owned, cul- corporate nutritional guidelines for the congregation and turally anchored, and centred on a collective approach. another [63] invested in five members to become key fa- Culturally relevant interventions are essential for uptake cilitators for the nutrition and physical activity compo- and maintenance of behaviour change programmes. nents. Churches can establish themselves in a way where CBPR provides a useful framework to ensure interven- health interventions could be mobilised with the right tions are culturally grounded, which is vital for the up- support structures in place. Churches have long been take and maintenance of behaviour change when viewed as an extension of the family, preserving tradi- programmes initially intended for non-Pacific popula- tions and cultures, and mediating between the commu- tions are adapted. Framing behaviour change from the nity and broader society [22]. As such, it is not context of Pacific cultural values becomes an integral surprising positive behaviour change outcomes were part of this process. Moreover, negotiating these spaces reported within this faith-based setting. through talanoa and understanding how the physical,
Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 13 of 15
social, and spiritual elements are intrinsically linked to Received: 6 August 2020 Accepted: 10 February 2021
the sustenance of Pacific people’s health and wellbeing is
critical [28]. While interventions can include common
cultural elements; approaches need to be contextualised References
to each Pacific Island community. 1. Hunter DJ, Reddy KS. Noncommunicable diseases. N Engl J Med. 2013;
Community centred aspirations determined by Pacific 36914:1336–43. https://doi.org/10.1056/NEJMra1109345.
2. Tolley H, Snowdon W, Wate J, Durand AM, Vivili P, McCool J, et al.
communities is fundamental to ensuring the health out- Monitoring and accountability for the Pacific response to the non-
comes measured by interventions, are elements that are communicable diseases crisis. BMC Public Health. 2016;161:958–12. https://
relevant and applicable to their lived realities and world- doi.org/10.1186/s12889-016-3614-8.
3. Pacific Perspectives Limited. Tofa Saili: A review of evidence about health
views. Future research needs to invest in building equity for Pacific Peoples in New Zealand. Wellington: Pacific Perspectives;
research capacity within Pacific communities, centering 2019.
self-determining research agendas, and findings to be led 4. World Health Organisation. Addressing noncommunicable diseases in the
Pacific islands, https://www.who.int/westernpacific/activities/addressing-
and owned by communities. This recognises Pacific ncds-in-the-pacific (2020).
communities are more than programme facilitators. 5. Ministry of Health. Health and Independence Report 2018. Wellington:
Ministry of Health; 2020.
6. Ministry of Health. Tupu Ola Moui: Pacific health chart book 2012.
Supplementary Information Wellington: Ministry of Health; 2012.
The online version contains supplementary material available at https://doi. 7. Ministry of Health and Ministry of Pacific Island Affairs. Tupu Ola Moui:
org/10.1186/s12889-021-10420-9. Pacific health chart book 2004. Wellington: Ministry of Health and Ministry
of Pacific Island Affairs; 2004.
Additional file 1. PRISMA study selection process 8. Ministry of Health. Annual data explorer 2017/18. New Zealand: Health
Survey; 2019. https://www.health.govt.nz/publication/annual-update-key-
results-2017-18-new-zealand-health-survey.
Abbreviations 9. Hills AP, Byrne NM, Lindstrom R, Hill JO. Small changes' to diet and physical
BCT: Behavior change techniques; CBPR: Community based participatory activity behaviors for weight management. Obes Facts. 2013;63:228–38.
research; CCF: Community coach facilitators; CHW: Community health https://doi.org/10.1159/000345030.
workers; HbA1c: Haemoglobin A1c; HCA: Health care assistant; PHE: Peer 10. Riekert K, Ockene J, Pbert L. The handbook of health behaviour change. 4th
health educators; POP: Pili Ohana Project; PRISMA: Preferred Reporting Items ed. New York: Springer Publishing Company; 2014.
for Systematic Reviews and Meta-Analyses; SCT: Social cognitive theory; 11. Michie S, Johnson M. Behaviour change techniques. In: Gellman MD, Rick
USA: United States of America Turner J, editors. Encyclopaedia of Behavioral medicine. New York: Springer;
2013. p. 182.
Acknowledgements 12. Glanz K, Bishop DB. The role of behavioral science theory in development
Not applicable and implementation of public health interventions. Annu Rev Public
Health. 2010;311:399–418. https://doi.org/10.1146/annurev.publhealth.012
Authors’ contributions 809.103604.
AI designed the search strategy, reviewed by JM and RD. AI conducted the 13. Davis R, Campbell R, Hildon Z, Hobbs L, Michie S. Theories of behaviour and
search, screening, data-extraction, and analysis. JM and RD assisted with data behaviour change across the social and behavioural sciences: a scoping
interpretation. AI drafted the manuscript and all authors contributed to revi- review. Health Psychol Rev. 2015;93:323–44. https://doi.org/10.1080/174371
sions, critically reviewing manuscript drafts, reading, and approving the final 99.2014.941722.
manuscript. 14. Tukuitonga C. Impact of culture on health. Pacific Health Dialog. 2018;211:
5–7. https://doi.org/10.26635/phd.2018.901.
Funding 15. Tiatia-Seath J. The importance of Pacific cultural competency in healthcare.
This research was funded by the Health Research Council of New Zealand Pacific Health Dialog. 2018;211:8–9. https://doi.org/10.26635/phd.2018.909.
Pacific PhD Scholarship. The funder had no role in designing the study, 16. Health Research Council of New Zealand. Pacific Health Research
collecting, analysing, or interpreting data, or in writing the manuscript. Guidelines. Auckland: Health Research Council of New Zealand; 2014.
17. Ministry for Pacific Peoples. Yavu: Foundations of Pacific development.
Availability of data and materials Wellington: Ministry for Pacific Peoples; 2018.
The datasets used and analysed in the current study are available from the 18. Grey A. The definition and measurement of ethnicity: a Pacific perspective.
corresponding author on reasonable request. Wellington: Gray Matter Research Ltd; 2001.
19. Ministry of Health. Making a Pacific difference: Strategic initiatives for
the health of Pacific people in New Zealand. Wellington: Ministry of
Ethics approval and consent to participate
Health; 1997.
Not applicable.
20. Fa'alau F, Jenson V. Samoan youth and family relationships in Aotearoa New
Zealand. Pacific Public Health. 2006;132:17–24.
Consent for publication
21. Norris P, Fa'alau F, Va'ai C, Churchward M, Arroll B. Navigating between
Not applicable.
illness paradigms: Treatment seeking by Samoan people in Samoa and New
Zealand. Qual Health Res. 2009;1910:1466–75. https://doi.org/10.1177/104
Competing interests 9732309348364.
The authors declare that they have no competing interests. 22. Ministry of Health. Improving quality of care for Pacific peoples. Wellington:
Ministry of Health; 2008.
Author details 23. Pulotu-Endemann FK. Fonofale model of health. Paper presented at the
1
Epidemiology and Biostatistics, School of Population Health, University of paper presented at the Pacific models for health promotion workshop,
Auckland, Private Bag 92019, Auckland 1142, New Zealand. 2Moana Research, Massey University. Wellington, New Zealand. https://whanauoraresearch.co.
Nga Hau Māngere Birthing Centre, 14 Waddon Place, Auckland, Māngere, nz/wp-content/uploads/formidable/Fonofalemodelexplanation1-Copy.pdf.
New Zealand. 3National Institute for Health Innovation, University of 24. Lee H, Francis ST. Migration and transnationalism: Pacific perspectives:
Auckland, Auckland, New Zealand. 4Pacific Health Section, School of Australian National University Press; 2009. p. 7.
Population Health, University of Auckland, Auckland, New Zealand.Matenga-Ikihele et al. BMC Public Health (2021) 21:435 Page 14 of 15
25. Mavoa HM, McCabe M. Sociocultural factors relating to Tongans' and Native Hawaiian-serving worksites in Hawai'i. Transl Behav Med. 2016;62:
indigenous Fijians' patterns of eating, physical activity and body size. Asia 190–201. https://doi.org/10.1007/s13142-015-0383-3.
Pac J Clin Nutr. 2008;173:375. 48. Ing CT, Zhang G, Dillard A, Yoshimura SR, Hughes C, Palakiko DM, et al.
26. Sheridan N, Kenealy T, Salmon E, Rea H, Raphael D, Schmidt-Busby J. Social support groups in the maintenance of glycemic control after
Helplessness, self-blame and faith may impact on self-management in community-based intervention. J Diabetes Res. 2016;2016:7913258. https://
COPD: a qualitative study. Prim Care Respir J. 2011;203:307–14. https://doi. doi.org/10.1155/2016/7913258.
org/10.4104/pcrj.2011.00035. 49. Mau MK, Glanz K, Severino R, Grove JS, Johnson B, Curb JD. Mediators of
27. Ryan D, Southwick M, Teevale T, et al. Primary care for Pacific people. lifestyle behavior change in native Hawaiians: initial findings from the native
Wellington: Pacific Perspectives; 2011. Hawaiian diabetes intervention program. Diabetes Care. 2001;2410:1770–5.
28. Firestone R, Funaki T, Dalhousie S, Henry A, Vano M, Grey J, et al. Identifying https://doi.org/10.2337/diacare.24.10.1770.
and overcoming barriers to healthier lives. Pacific Health Dialog. 2018;212: 50. Beckham S, Bradley S, Washburn A, Taumua T. Diabetes management:
54–66. https://doi.org/10.26635/phd.2018.913. utilizing community health workers in a Hawaiian/Samoan population. J
29. Sa'uLilo L, Tautolo E, Egli V, Smith M. Health literacy of Pacific mothers in Health Care Poor Underserved. 2008;192:416–27. https://doi.org/10.1353/
New Zealand is associated with sociodemographic and non-communicable hpu.0.0012.
disease risk factors: surveys, focus groups and interviews. Pacific Health 51. Gellert KS, Aubert RE, Mikami JS, Ola K'A. Moloka'i's community-based
Dialog. 2018;212:65–70. https://doi.org/10.26635/phd.2018.914. healthy lifestyle modification program. Am J Public Health. 2010;1005:779–
30. Baig AA, Benitez A, Quinn MT, Burnet DL. Family interventions to improve 83. https://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2009.176222.
diabetes outcomes for adults. Ann N Y Acad Sci. 2015;13531:89–112. https:// 52. Tomioka M, Braun KL, Compton M, Tanoue L. Adapting Stanford's chronic
doi.org/10.1111/nyas.12844. disease self-management program to Hawaii's multicultural population.
31. Human Rights Commission. Human Rights in New Zealand. Auckland: Gerontologist. 2012;521:121–32. https://doi.org/10.1093/geront/gnr054.
Human Rights Commission; 2010. 53. Kaholokula J, Look M, Mabellos T, Zhang G, de Silva M, Yoshimura S, et al.
32. Paddison CAM. Exploring physical and psychological wellbeing among Cultural dance program improves hypertension management for native
adults with type 2 diabetes in New Zealand: identifying a need to improve Hawaiians and Pacific islanders: a pilot randomized trial. J Racial Ethn Health
the experiences of Pacific peoples. N Z Med J. 2010;1231310:30. Disparities. 2017;41:35–46. https://doi.org/10.1007/s40615-015-0198-4.
33. Southwick M, Kenealy T, Ryan D. Primary care for Pacific people: a Pacific 54. Ing CT, Miyamoto RES, Fang R, Antonio M, Paloma D, Braun KL, et al.
and health systems approach. Wellington: Pacific Perspectives; 2012. Comparing weight loss–maintenance outcomes of a worksite-based lifestyle
34. Tualaulelei E, McFall-McCaffery J. The Pacific research paradigm : program delivered via DVD and face-to-face: a randomized trial. Health
opportunities and challenges. MAI J (Online). 2019;82:188–204. https://doi. Educ Behav. 2018;454:569–80. https://doi.org/10.1177/1090198118757824.
org/10.20507/MAIJournal.2019.8.2.7. 55. Mau MKLM, Lim E, Kaholokula JK, TMU L, Cheng Y, Seto TB. A randomized
35. Anae M. Teu Le Va: Samoan relational ethics. Knowl Cult. 2016;43:117–30. controlled trial to improve heart failure disparities: The Mālama Pu‘uwai
36. Yang K, Chung-Do JJ, Fujitani L, Foster A, Mark S, Okada Y, et al. Advancing (caring for heart) Study. Open Access J Clin Trials. 2017;9:65–74. https://doi.
community-based participatory research to address health disparities in org/10.2147/OAJCT.S136066.
Hawaii: Perspectives from academic researchers. Hawai‘i J Med Public 56. LaBreche M, Cheri A, Custodio H, Fex CC, Foo MA, Lepule JT, et al. Let's
Health. 2019;783:83–7. move for Pacific Islander communities: an evidence-based intervention to
37. Hudson M, Milne M, Reynolds P, et al. Te Ara Tika guidelines for Māori increase physical activity. J Cancer Educ. 2016;312:261–7. https://doi.org/10.1
research ethics: a framework for researchers and ethics committee 007/s13187-015-0875-3.
members the Pūtaiora writing group. Auckland: Health Research Council of 57. McElfish PA, Bridges MD, Hudson JS, Purvis RS, Bursac Z, Kohler PO, et al.
New Zealand; 2010. Family model of diabetes education with a Pacific Islander community.
38. Australian Institute of Aboriginal and Torres Strait Islander Studies. AIATSIS Diabetes Educ. 2015;416:706–15. https://doi.org/10.1177/0145721715606806.
code of ethics for Aboriginal and Torres Strait Islander research. 58. DePue JD, Dunsiger S, Seiden AD, Blume J, Rosen RK, Goldstein MG, et al.
Canberra. Australian Institute of Aboriginal and Torres Strait Islander Nurse-community health worker team improves diabetes care in American
Studies (AIATSIS). 2020. Samoa: results of a randomized controlled trial. Diabetes Care. 2013;367:
39. Assembly of First Nations. Ethics in First Nations research. Canada: 1947–53. https://doi.org/10.2337/dc12-1969.
Environment Stewardship Unit; 2009. 59. Hamid S, Dunsiger S, Seiden A, Nu'Usolia O, Tuitele J, DePue JD, et al.
40. Mila-Schaaf K, Hudson M. The interface between cultural understandings: Impact of a diabetes control and management intervention on health care
negotiating new spaces for Pacific mental health. Pacific Health Dialog. utilization in American Samoa. Chronic Illness. 2014;102:122–34. https://doi.
2009;151:113. org/10.1177/1742395313502367.
41. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for 60. Reddy R, Trinidad R, Seremai J, Nasa J. Marshallese diabetic health
systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. improvement pilot project in Ebeye. Californian J Health Promot. 2009;7:
2009;67:e1000097. https://doi.org/10.1371/journal.pmed.1000097. 125–30. https://doi.org/10.32398/cjhp.v7iSI.2006.
42. Joanna Briggs Institute. The Joanna Briggs Institute Critical Appraisal tools 61. Tan J, McCready F, Noovao N, Tepueluelu O, Collins J, Cundy T.
for use in JBI systematic Reviews: Checklist for Randomized Controlled Trials; Intensification of blood pressure treatment in Pasifika people with type 2
2017. https://doi.org/10.1371/journal.pone.0208900.t002. diabetes and renal disease: a cohort study in primary care. N Z Med J. 2014;
43. Mau MK, Kaholokula JK, West MR, Leake A, Efird JT, Rose C, et al. Translating 1271404:17–26. https://pubmed.ncbi.nlm.nih.gov/25331308/.
diabetes prevention into native Hawaiian and Pacific Islander communities: 62. Simmons D, Voyle JA, Fou F, Feo S, Leakehe L. Tale of two churches:
the PILI 'Ohana Pilot project. Prog Community Health Partnersh. 2010;41:7– differential impact of a church-based diabetes control programme among
16. https://doi.org/10.1353/cpr.0.0111. Pacific Islands people in New Zealand. Diabet Med. 2004;212:122–8. https://
44. Kaholokula JK, Mau MK, Efird JT, Leake A, West M, Palakiko DM, et al. A doi.org/10.1111/j.1464-5491.2004.01020.x.
family and community focused lifestyle program prevents weight regain in 63. Bell AC, Swinburn BA, Amosa H, Scragg RK. A nutrition and exercise
Pacific Islanders: a pilot randomized controlled trial. Health Educ Behav. intervention program for controlling weight in Samoan communities in
2012;394:386–95. https://doi.org/10.1177/1090198110394174. New Zealand. Int J Obes Relat Metab Disord. 2001;256:920. https://doi.org/1
45. Sinclair KA, Makahi EK, Shea-Solatorio C, Yoshimura SR, Townsend CK, 0.1038/sj.ijo.0801619.
Kaholokula JK. Outcomes from a diabetes self-management intervention for 64. Simmons D, Fleming C, Cameron M, Leakehe L. A pilot diabetes awareness
Native Hawaiians and Pacific People: Partners in Care. Ann Behav Med. and exercise programme in a multiethnic workforce. N Z Med J. 1996;
2013;451:24–32. https://doi.org/10.1007/s12160-012-9422-1. 1091031:373–5.
46. Kaholokula J, Kaholokula J, Wilson R, Wilson R, Townsend C, Townsend C, 65. Hotu C, Bagg W, Collins J, Harwood L, Whalley G, Doughty R, et al. A
et al. Translating the diabetes prevention program in native Hawaiian and community-based model of care improves blood pressure control and
Pacific islander communities: the PILI ‘Ohana project. Transl Behav Med. delays progression of proteinuria, left ventricular hypertrophy and diastolic
2014;42:149–59. https://doi.org/10.1007/s13142-013-0244-x. dysfunction in Maori and Pacific patients with type 2 diabetes and chronic
47. Ing CK, Miyamoto RE, Antonio M, Zhang G, Paloma D, Basques D, et al. The kidney disease: a randomized controlled trial. Nephrol Dial Transplant. 2010;
PILI@Work Program: a translation of the diabetes prevention program to 2510:3260–6. https://doi.org/10.1093/ndt/gfq168.You can also read
