Is Neck Posture Subgroup in Late Adolescence a Risk Factor for Persistent Neck Pain in Young Adults? A Prospective Study

Abstract

Objective

The purpose of this study was to determine whether sagittal neck sitting posture subgroup membership in late adolescence was a risk factor for persistent neck pain (PNP) in young adults.

Methods

There were 686 participants enrolled in the Raine Study at the 17- and 22-year follow-ups. At 17 years of age, posture was measured by photographs, and 4 subgroups of sitting neck posture were determined by cluster analysis. Height and weight were measured, and exercise frequency, depression, and PNP were assessed by questionnaire. At 22 years of age, participants answered questions about neck pain and occupation type. Logistic regression examined if neck posture subgroups at 17 years of age were a risk factor for PNP at 22 years of age, taking into account other factors.

Results

Female sex (odds ratio [OR] = 1.75, 95% CI = 1.16–2.65) and PNP at 17 years of age (OR = 3.78, 95% CI = 2.57–5.57) were associated with PNP at 22 years of age. In females, neck posture subgroup at 17 years of age was a risk factor for PNP at 22 years of age. Compared with the upright subgroup, both the slumped thorax/forward head subgroup groups and the intermediate subgroup had decreased odds for PNP at 22 years of age (OR = 0.24, 95% CI = 0.08–0.76; OR = 0.38, 95% CI = 0.15–0.99, respectively). No association was found in males.

Conclusion

After taking into account PNP at 17 years, sitting neck posture at 17 was not a risk factor for PNP at 22 years of age in males, whereas in females, more relaxed postures (slumped thorax/forward head and Intermediate postures) were protective of neck pain compared with upright posture.

Impact

Females in late adolescence who sat in slumped thorax/forward head or intermediate posture rather than upright sitting posture had a lower risk of PNP as a young adult. The practice of generic public health messages to sit up straight to prevent neck pain needs rethinking.

Posture, Neck Pain, Risk Factors, Adolescent, Young Adult, Raine Study

Introduction

Neck pain (NP) is the fourth-ranked disorder for numbers of years lived with disability worldwide.¹ NP often originates in childhood,² with prevalence reported to increase through adolescence, peak in middle age, and decline thereafter.³ NP is a complex, multifactorial condition with a range of different reported risk factors, including female sex^4–7 small height,⁸ psychological distress/poorer mental health,^4,9 past history of NP,⁴ lower exercise participation,^5,10,11 and occupational factors such as computer use.¹²

Posture is commonly believed to be both associated with and a risk factor for NP.^13–17 It has been proposed that both very upright and very protracted neck postures could contribute to cumulative biomechanical load on the spine, which may sensitize articular and myofascial structures and result in pain.¹⁸ However, 3 systematic reviews in adult and adolescent populations reported that cross-sectional evidence does not support that sagittal spinal alignment differs between those with or without NP.^19–21 A recent systematic review with meta-analysis reported that age was a confounding factor in the association between neck posture and NP and that adults with NP have a more forward head position than those without, whereas no such association is found in adolescents.²² Previous studies have been limited by partial characterization of posture, often with singular measures of neck posture without considering the influence of thorax and head position,^23,24 and lack of control for other factors potentially associated with NP.¹⁹ Addressing these limitations, a recent cross-sectional study characterized sagittal neck posture using 5 posture measures and identified 4 separate subgroups of sitting neck posture in 1100 17-year olds.²⁵ When accounting for multiple factors (sex, height, weight, exercise frequency, depression), no association was found between neck posture subgroups and NP.

Prospective studies examining posture as a risk factor for NP are scant. One study of 1334 adult workers without NP did not identify neck flexion as a risk factor for incident NP over 3 years, even after consideration of time spent in neck flexion and biopsychosocial factors that could confound the relationship.²⁶ However, only one-quarter of the workers in each occupational group was observed, and a high proportion of male participants were enrolled in the study; selection bias may account for the lack of association. Another prospective study found adolescents whose posture was at the extremes of neck and thoracic angles (ie, both upright and flexed postures) were at increased risk for upper quadrant pain.²⁷ However, this study had a small sample size (n = 93) and because 10 areas in the upper quadrant were considered rather than NP specifically, it does not provide direct evidence for a link between posture and NP. However, the results of that study suggest extremes of posture (ie, closer to end range extension or flexion) may be a risk factor for NP.

Despite NP being prevalent in adolescence,²⁸ there is a lack of longitudinal studies examining posture while considering other risk factors for NP in young adults that might confound this association. This study aimed to determine if subgroups of sitting neck posture identified at 17 years of age were risk factors for NP at 22 years independent of other potential risk factors.

Methods

Participants

Participants were enrolled in the Raine Study (www.rainestudy.org.au). The Raine Study design has been described previously.²⁹ In brief, the Raine Study commenced in 1989 with a predominantly Caucasian cohort of 2868 participants who were enrolled at approximately their 18th week of gestation. The participants have been followed-up at regular intervals through their lives. The characteristics of the active participants at age 22 years have been compared with census data, and the sample is representative of similarly aged young adults in Western Australia.³⁰ The current study utilized de-identified posture and questionnaire data at the 17-year follow-up and questionnaire data at the 22-year follow-up. Prior to participation, informed written consent was obtained from all participants at 22 and 17 and from their parents at 17 years. The study received ethical approval from the Curtin University Human Research Ethics Committee and the Human Research Ethics Committees at the University of Western Australia.

Measures

Postural Assessment

A subgroup of the Raine Study participants provided postural data at 17 years. Sitting sagittal neck posture was assessed by photographs. Four subgroups, termed upright, intermediate, slumped thorax/forward head, and erect thorax/forward head (Fig. 1), were derived from a composite of 5 posture measures using methods previously described.²⁵ Measures were selected as they represented sagittal postural measures commonly assessed in clinical practice and have been proposed to be a biomechanical mechanism for NP.^31,32

Figure 1

Characteristics of neck posture subgroups.

Neck Pain

Neck pain questions at age 17 and 22 years were adapted from the Nordic questionnaire, for which validity and reliability evidence has been reported.³³ Participants answered “yes or no” to “Have you ever had neck pain?”; “Has your neck pain ever lasted more than 3 months continuously (it hurt more or less every day)?”; and “Has your neck pain ever lasted for 3 months on and off (it hurt once a week but not every day)?” A positive response from either of the latter 2 questions was considered to indicate the presence of persistent neck pain (PNP).

Anthropometric Factors

Height and weight were measured at 17 years using standard methods.

Psychological Factors

The Beck Disability for Youth was used to assess depression at the age of 17.³⁴ Good validity and reliability of the Beck Disability for Youth has been reported.^34,35 Scores were dichotomized as recommended for clinical classification of depression into those with no or minimal depression and those with mild, moderate, or severe depression.³⁶

Lifestyle Factors

At age 17, participants were asked: “Outside school, Technical and Further Education (TAFE) or work hours how often do you exercise in your free time, so much so you get out of breath or sweat: once a month or less; once a week; 2 to 3 times a week; 4 to 6 times a week/every day?” Responses were dichotomized to those who undertook exercise once per week or less vs those who exercised more than once per week in line with current exercise guidelines.³⁷

Work Physical Demands

At age 22, participants were asked; “Which of the following statements best describes the work that you do in your current job? Sedentary occupation (eg, secretary: where you spend most of your time sitting); standing occupation (eg, shop assistant, security guard: spend most of your time standing/walking but not intense physical effort); physical work (eg, plumber, nurse: a job that requires some physical effort including handling of heavy objects and use of tools); or heavy manual work (eg, bricklayer: a job that involves very vigorous physical activity including handling very heavy objects).” Questions were derived from a nutrition and physical activity questionnaire, for which the validity has been reported.³⁸

Analysis

Associations between covariates of interest at age 17 and PNP at 22 years, as well as occupation at 22 years, were analyzed using chi-squared tests (categorical variables) or linear regression (continuous variables). A series of logistic regression models was used to assess the association between posture clusters and PNP adjusted for confounding. Initially, univariable logistic regression was used to estimate the unadjusted odds ratios (OR) for PNP referenced to no PNP at 22 years for posture cluster membership and each potential confounder of interest. Then, posture cluster membership and potential confounding variables that were significant univariably at the P < .10 level were included in an initial adjusted model (Model 1), and potential confounding variables that were not significant at P < .10 in Model 1 were excluded in a final adjusted model (Model 2). It was hypothesized that males and females may have different associations between posture and PNP at 22; therefore, an interaction term between sex and posture subgroup membership was assessed to investigate if the association between posture subgroup membership with PNP at 22 years varied by sex. It was also hypothesized that participants with and without PNP at 17 years may have different associations between posture and NP; therefore, an interaction between PNP at 17 years and posture clusters were also assessed to check equivalence of the association between posture subgroup membership with PNP at 22 between those with and without PNP at 17 years. Statistical analysis was performed using Stata version 13 for Mac (StataCorp LP, College Station, TX, USA) with α < .05.

Role of the Funding Source

The core management of the Raine Study is funded by The University of Western Australia (UWA), Raine Medical Research Foundation, Telethon Kids Institute, Women and Infants Research Foundation, Curtin University, Murdoch University, The University of Notre Dame Australia and Edith Cowan University. The 17-year Raine Study follow-up was funded by the National and Medical Research Council (NHMRC) program grant 353514 and NMHRC project grant 323200. The 22-year Raine Study follow-up was funded by NHMRC project grants 1027449, 1044840, and 102185. Funding was also generously provided by SafeWork Australia. These funding sources had no involvement in the design, conduct, or reporting in this study.

Results

Participant flow through the study is shown in Figure 2. There were no significant differences in the characteristics of the analysis sample at 22 years compared with those who did not return for follow-up (Tab. 1). Missing data on covariates of interest are indicated in Table 2. At 22 years, 192 of 686 participants (28%) reported PNP, which had risen significantly from 152 of 686 participants (22%) with PNP at 17 years (P ≤ .001). Persistent NP at 17 years was strongly associated with PNP at 22 years, with a significant fourfold increase in odds of PNP at 22 years if participants had experienced PNP at 17 years (OR = 4.19, 95% CI = 2.86–6.12, P < .001; Tab. 2).

Figure 2

Flow of participants.

Table 1

Sample at Age 22 Years Compared With Those Who Did Not Attend Follow-Up

	Analysis Sample n = 686	Not at Follow-Up n = 289	P
Posture subgroup			.776a
Upright	204 (29%)	82 (28%)
Intermediate	156 (23%)	74 (26%)
Slumped thorax/forward head	103 (15%)	45 (15%)
Erect thorax/forward head	223 (32%)	88 (30%)
PNP at 17	152 (22%)	65 (22%)	.909 a
Female sex	365 (53%)	135 (47%)	.064 a
Depression (mild or more)	143 (21%)	68 (24%)	.347 a
Exercise (>1x/wk)	413 (61%)	157 (57%)	.198 a
Height (m)	1.72 (1.71–1.72)	1.72 (1.71–1.73)	.483b
Weight (kg)	66.8 (65.8–67.8)	66.9 (65.4–68.6)	.846 b

^aAnalyzed with chi-squared test. PNP = persistent neck pain.

^bAnalyzed with t test.

Table 2

Prevalence of Neck Pain at 22 Years by Posture Cluster Membership and Covariates of Interest

	N	PNP at 22 (28.0%)	No PNP at 22 (72%)	P
Posture subgroup n, (%)				.002a
Upright	204	72 (35.3%)	132 (64.7%)
Intermediate	156	36 (23.1%)	120 (76.9%)
Slumped thorax/forward head	103	17 (16.5%)	86 (83.5%)
Erect thorax/forwardhead	223	67 (30.0%)	156 (70%)
PNP at 17 y, n (%)				<.001a
No	534	112 (21.0%)	422 (79.0%)
Yes	152	80 (52.6%)	72 (47.4%)
Sex n, (%)				<.001a
Male	321	63 (19.7%)	258 (81%)
Female	365	129 (35.3%)	236 (65%)
Depression,b n (%)				.078a
No	527	141 (26.8%)	386 (73.2%)
Yes (mild or more)	143	49 (34.3%)	94 (65.7%)
Exercise >1x/wk,c n (%)				.436a
No	262	77 (29.4%)	185 (70.6%)
Yes	413	110 (26.6%)	303 (73.3%)
Occupation type,d n (%)				.575a
Sedentary	168	48 (28.6%)	120 (71.4%)
Standing	205	63 (30.7%)	142 (69.3%)
Physical	163	40 (24.5%)	123 (75.5%)
Heavy manual work	33	8 (24.2%)	25 (75.7%)
Anthropometric factors: mean, (SD)
Height (m)		1.70 (0.09)	173 (0.09)	.004
Weight (kg)e		66.1 (14.3)	67.8 (13.7)	.005

^aAnalyzed with chi-squared test. PNP = persistent neck pain.

^bMissing for 16 cases.

^cMissing for 11 cases.

^dMissing for 117 cases.

^eMissing for 1 case.

Table 2 displays the prevalence of PNP at 22 years across categorical covariates of interest. In univariable analysis, posture subgroup was associated with PNP at 22 years (P = .002; Tab. 3). Separate results for the association between PNP and posture subgroups are reported for males and females because there was evidence for interaction between sex and posture subgroups at the P < .1 level (P = .057).

Table 3

Univariable and Multivariable Logistic Regression Models for PNP at 22 Years

	Unadjusted			Model 1a			Model 2b
	OR	95% CI	P	OR	95% CI	P	OR	95% CI	P
Posture Subgroup
Males
Upright	REF		.668	REF		.479	REF		0.537
Intermediate	1.58	0.69, 3.59	.278	1.83	0.78, 4.3	.162	1.77	0.76, 4.15	0.186
Slumped thorax/forward head	1.13	0.45, 2.88	.787	1.22	0.47, 3.2	.683	1.21	0.46, 3.15	0.698
Erect thorax/forward head	1.25	0.49, 3.24	.639	1.33	0.49, 3.63	.566	1.43	0.54, 3.80	0.474
Females
Upright	REF		.016	REF		.0376	REF		0.023
Intermediate	0.42	0.17, 1.06	.065	0.37	0.14, 0.97	.042	0.38	0.15, 0.99	0.047
Slumped thorax/forward head	0.21	0.07, 0.64	.006	0.27	0.09, 0.87	.028	0.24	0.08, 0.76	0.015
Erect thorax/forward head	0.70	0.44, 1.08	.128	0.68	0.41, 1.12	.130	0.67	0.41, 1.09	0.109
PNP 17	4.42	2.90, 6.19	<.001	3.66	2.46, 5.45	<.001	3.85	2.60, 5.71	<0.001
Height (cm)	0.97	0.95, 0.99	.004	1.00	0.97, 1.03	.826
Weight (kg)	0.99	0.97, 1.00	.060	1.00	0.98, 1.01	.901
Depression	1.43	0.96, 2.12	.078	1.15	0.74, 1.78	.524
Exercise	0.87	0.62, 1.23	.436
Occupation type
Sedentary	REF		.572
Standing	1.10	0.71, 1.73	.650
Physical	0.81	0.5, 1.33	.407
Heavy manual	0.8	0.34, 1.9	.613

^aAdjusted for covariates with univariable association P < .10. PNP = persistent neck pain.

^bAdjusted for covariates significant in Model 1 at P < .10.

Although PNP at 22 years was more prevalent in those participants with mild or greater symptoms of depression at 17 years (Tab. 2), this association was not significant (OR = 1.43 95%, CI = 0.96–2.12, P = .078; Tab. 3). There was no difference in PNP between those participants reporting exercising more or less than once per week at 17 years (Tabs. 2 and 3). Taller participants at 17 years were found to have significantly lower odds for PNP at 22 years (OR = 0.97 per cm increase, 95% CI = 0.95–0.99, P = .004; Tab. 3). Heavier participants at 17 years had slightly lower odds for PNP at 22 years, although this association was not statistically significant (OR = 0.99 per kg increase, 95% CI = 0.97–1.00, P = .06; Tab. 3). There was no evidence that work physical demands at 22 years were associated with PNP at 22 years (P = .57; Tab. 3).

An initial multivariable model indicated that only PNP at 17 years remained independently and significantly associated with the presence of PNP at 22 years (Model 1, Tab. 3); thus, it was included in the final model to assess the association of posture subgroup membership with PNP at 22 years adjusting for confounding (Model 2, Tab. 3). In this final adjusted model, the overall test for significance for differences in odds between subgroups for the presence of PNP at 22 years was not significant in males (P = .54; Tab. 3) but was significant in females (P = .02; Tab. 3).

In females, the estimates of the prevalence of PNP at 22 years by subgroup adjusted for PNP at 17 years were highest in upright (40.1%) and lowest in slumped thorax/forward head (15.6%), and 21.8% in intermediate and 32.1% in erect thorax/forward head. Compared with upright, both the slumped thorax/forward head and the intermediate subgroups had decreased odds for the presence of PNP at 22 years (OR = 0.24, 95% CI = 0.08–0.76, P = .015; OR = 0.38, 95% CI = 0.15–0.99, respectively; P = .047; Tab. 3), but there was no evidence for decreased odds in the erect thorax/forward head subgroup (OR = 0.67, 95% CI = 0.41–1.09, P = .11).

There was no statistical evidence that the difference in the odds for PNP at 22 years in the 4 subgroups differed according to the presence or absence of PNP at 17 years in either males (P = .75) or females (P = .64).

Discussion

To our knowledge, this is the first large longitudinal study to examine the relationship between posture subgroups and NP in young people. A key finding of this study was that posture subgroup membership at 17 years was not a risk factor for PNP at 22 years in males but was in females, after adjusting for previous PNP. A novel finding of this study is that sitting neck posture subgroups carries a different risk of NP between males and females. A previous workplace study found that neck flexion did not increase the risk of incident NP over 3 years; however, 75% of the participants enrolled in the study were male and the results were not stratified by sex, which may account for the difference in findings to this study.²⁶ In contrast, 2 small studies reported extremes of neck, neck and thoracic angles,²⁷ and head flexion³⁹ were a risk factor for 10 areas of upper quadrant pain in adolescents using computers but not specifically NP. In contrast to the current study, males with extreme upright or flexed neck and thoracic angles had a greater risk of upper quadrant musculoskeletal pain than females in 1 study,²⁷ but sex was not found to confound the relationship between pain and head flexion in another.³⁹ Methodological differences may underpin these contrary findings, related to the measure of 10 areas of pain in the upper quadrant, rather than NP, a small sample size, and that neck posture angles were measured in relation to computer use.⁴⁰

Use of sitting posture subgroups and the sex-specific analysis in the current study revealed a high prevalence of PNP in females who sat in upright posture compared with a low prevalence of PNP in females who sat in the opposite extreme of posture, slumped thorax/forward head. The current study findings may help to explain why systematic reviews found no cross-sectional relationship between sagittal spinal alignment and NP.^19–21 Previous studies have been limited by a lack of postural subgrouping and/or accounting for other related factors. Furthermore, school-based and workplace intervention studies reported no effect on NP following generic posture interventions.^41–43 Of note, our previous work found no cross-sectional relationship between neck posture subgroups and NP at 17 years in this cohort, despite adjusting for sex and other related factors.²⁵ This may be explained by the higher prevalence of PNP at follow-up, which revealed differences between males and females across the 4 different postural subgroups at 22 years, which was not evident at 17 years.

This study found PNP at 17 years was a risk factor for PNP at 22 years, which is consistent with prior evidence that the presence of NP in 15- to 18-year-old adolescents was associated with a high 6-month prevalence of NP in 22- to 25-year olds.⁵ This finding is consistent with reports for low back pain being predictive of future low back pain in the same cohort⁴⁴ as well as in adult populations.⁴⁵ It has been proposed that this phenomenon likely reflects adaptive changes within the peripheral and central nervous processes resulting in sensitisation.^45,46

Analogous with cross-sectional^7,9,28,47 and prospective studies^4,5 in adolescent and adult populations, this study found a greater prevalence of NP in females than males. It has been proposed that physiological, hormonal, and psychosocial differences in pain sensitivity or expression between males and females⁴⁸ may underpin these differences.⁴⁹ The current study contributes to this evidence by clarifying the sex-specific role of posture as a risk factor.

Two of the subgroups in this study were characterized by forward head posture. It has been previously hypothesized that biomechanical load created by the head being more forward of the center of gravity may lead to a higher prevalence of NP in people with habitual forward head posture.⁵⁰ In contrast, our findings found there was no difference in the presence of NP between posture subgroups in males and that in females the lowest prevalence of NP was in the slumped thorax/forward head posture subgroup while the subgroup with the highest prevalence of NP was the upright posture. This finding is also contrary to the beliefs held by health practitioners and in society that sitting up straight is optimal whereas forward head postures are not optimal^49,51 and a cause of NP.^13–15

Although speculative, one explanation for this finding may be related to the fact that thoracic upright postures compared with a slumped thorax/forward head posture are associated with higher activity of the upper thoracic erector spinae muscles, which attach to the cervical spine and therefore might expose the cervico-thoracic spine to greater loading through their multi-segmental attachments.⁵² Alternatively, muscle pain of the upper thoracic erector spinae may develop as a consequence of sustained low force to maintain extended postures.⁵³ Another explanation for this finding may be that more upright postures are a consequence of psychosocial factors such as heightened anxiety or hypervigilance, which may also mediate NP.

Although we acknowledge that population studies may not be sensitive enough to detect how posture relates to individual circumstance, the sex differences found between neck posture and NP, combined with the absence of a cross-sectional relationship in the literature, questions the validity of generic postural advice to sit up straight to prevent NP. Other biopsychosocial factors not measured in this study such as physiological responses to stress⁴⁶ or altered pain processing secondary to previous neck injury⁴ may be more important in predicting NP. These findings have implications at a societal level for providing evidence-informed public health information and dispelling health myths regarding the common belief that NP is caused by slumped or forward head posture.^13–15 Along with evidence from school-based and workplace intervention studies,^41–43 these findings also have implications for the occupational and public health industry that places a focus on optimal postural alignment to prevent and treat NP. Although speculative, it may be that factors such as postural variability rather than a preconceived ideal posture are more important targets for care. These findings also have implications for clinicians treating patients with NP to consider other modifiable biopsychosocial drivers of pain rather than body posture. Recent research supports the role of biopsychosocial intervention for more effective care of NP.⁵⁴

Strengths of this study include its longitudinal design, use of well-defined posture subgroups, sex-specific analyses, and the consideration of a range of potential confounders. In terms of limitations, approximately 30% of participants did not attend follow-up. However, the analysis sample did not differ from those that did not attend follow-up and remained representative of the participants at 17 years on all baseline measures. Capture of posture measures by photography cannot detect precise anatomical alignment such as intervertebral translation and rotation, which may occur with alterations of posture⁵⁵ and may be important in the development of NP. In addition, participants sat using standardized instructions in a laboratory setting, providing a static view of posture, which may not be representative of their normal habitual posture or account for postural variation over time.⁵⁶ Stability of posture subgroup membership over time is unknown. This study did not consider the cause of NP, aggravating activities, the severity of NP, or levels of impairment of the participants. Although this study considered depression as a confounder, we acknowledge that other psychosocial factors that may confound the neck posture and NP relationship were not accounted for.

This study found that for females, sitting in upright posture at 17 years was a risk factor for the presence of PNP at 22 years. However, there was no evidence that sitting posture at 17 years was a risk factor for NP at 22 years in males. Additionally, a previous history of NP increased the risk of PNP at 22. The results question the practice of generic population-based posture education to sit up straight for the prevention of NP.

Author Contributions

Concept/idea/research design: K.V. Richards, D.J. Beales, A.L. Smith, P.B. O’Sullivan, L.M. Straker

Writing: K.V. Richards, D.J. Beales, A.L. Smith, P.B. O’Sullivan, L.M. Straker

Data collection: L.M. Straker

Data analysis: K.V. Richards, A.L. Smith, L.M. Straker

Project management: D.J. Beales, P.B. O’Sullivan, L.M. Straker

Fund procurement: L.M. Straker

Providing participants: L.M. Straker

Providing facilities/equipment: L.M. Straker

Consultation (including review of manuscript before submitting): D.J. Beales, P.B. O’Sullivan, L.M. Straker

Acknowledgments

The authors are grateful to the Raine Study participants and their families, and we thank the Raine Study for cohort coordination and data collection.

Funding

The core management of the Raine Study is funded by The University of Western Australia (UWA), Raine Medical Research Foundation, Telethon Kids Institute, Women and Infants Research Foundation, Curtin University, Murdoch University, The University of Notre Dame Australia and Edith Cowan University. The 17-year Raine Study follow-up was funded by the National and Medical Research Council (NHMRC) programme grant 353514 and NMHRC project grant 323200. The 22-year Raine Study follow-up was funded by NHMRC project grants 1027449, 1044840, and 102185. Funding was also generously provided by SafeWork Australia.

Ethics Approval

Prior to participation, informed written consent was obtained from all participants at 22 and 17 and from their parents at 17 years. The study received ethical approval from the Curtin University Human Research Ethics Committee and the Human Research Ethics Committees at the University of Western Australia.

Disclosures

No form of payment, honorarium, or grant was given to any of the authors to produce this manuscript. Professor Leon Straker is the outgoing scientific director of the Raine Study, and Professor Anne Smith has taken over this role in 2020, which may be a perceived conflict of interest. Dr Karen Richards, Dr Darren Beales, and Professor Peter O’Sullivan have no actual, perceived, or potential conflict of interest to declare.