Photorefractive keratectomy (PRK) versus laser assisted in situ keratomileusis (LASIK) for hyperopia correction

Abstract

Background

Hyperopia, or hypermetropia (also known as long‐sightedness or far‐sightedness), is the condition where the unaccommodating eye brings parallel light to a focus behind the retina instead of on it. Hyperopia can be corrected with both non‐surgical and surgical methods, among them photorefractive keratectomy (PRK) and laser assisted In situ keratomileusis (LASIK). There is uncertainty as to whether hyperopic‐PRK or hyperopic‐LASIK is the better method.

Objectives

The objectives of this review were to determine whether PRK or LASIK leads to more reliable, stable and safe results when correcting a hyperopic refractive error.

Search methods

We searched CENTRAL (which contains the Cochrane Eyes and Vision Group Trials Register) (The Cochrane Library 2012, Issue 2), MEDLINE (January 1950 to February 2012), EMBASE (January 1980 to February 2012), Latin American and Caribbean Literature on Health Sciences (LILACS) (January 1982 to February 2012), the metaRegister of Controlled Trials (mRCT) (www.controlled‐trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the WHO International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). There were no date or language restrictions in the electronic searches for trials. The electronic databases were last searched on 17 February 2012. When trials are included in the review we will search the reference lists of the studies included in the review for information about further trials. We will use the Science Citation Index to search for papers that cite any studies included in this review. We did not handsearch journals or conference proceedings specifically for this review.

Selection criteria

We planned to include only randomised controlled trials (RCTs) comparing PRK against LASIK for correction of hyperopia and then perform a sensitivity analysis of pre‐ and post‐millennial trials since this is the mid‐point in the history of both PRK and LASIK.

Data collection and analysis

We did not identify any studies that met the inclusion criteria for this review.

Main results

As no studies met the inclusion criteria for this review, we discussed the results of non‐randomised trials comparing hyperopic‐PRK with hyperopic‐LASIK.

Authors' conclusions

No robust, reliable conclusions could be reached, but the non‐randomised trials reviewed appear to be in agreement that hyperopic‐PRK and hyperopic‐LASIK are of comparable efficacy. High quality, well‐planned open RCTs are needed in order to obtain a robust clinical evidence base.

Plain language summary

Photorefractive keratectomy (PRK) versus laser assisted in situ keratomileusis (LASIK) for correction of long‐sightedness

Hyperopia (long‐sightedness or far‐sightedness) is the condition where the relaxed eye brings parallel light to a focus behind the retina instead of on it. In order to correct hyperopia a variety of surgical techniques can be applied including PRK and LASIK. There exists an uncertainty as to which technique provides more accurate, stable and safe results. As no randomised controlled trials were found that met the inclusion criteria, we could not find definite answers to these questions and therefore concluded that more research is required.

Background

Description of the condition

Hyperopia, or hypermetropia (also known as long‐sightedness or far‐sightedness), is the condition where the unaccommodating (i.e. relaxed) eye brings parallel light to a focus behind the retina instead of on it. The overall refractive power of the eye (i.e. the ability to converge parallel light to a focus) is a function of: a) its axial length and b) the refractive power of the cornea and the crystalline lens. In hyperopia, either because the axial length is too short (axial hyperopia) or the overall refractive power of the eye is too low (refractive hyperopia), parallel light comes to a focus behind the retina. 
 The prevalence of hypermetropia appears to vary with age, country and ethnic group. It has been postulated that from the age of six to 18 years the prevalence of hyperopia decreases with increasing age (Czepita 2005) and that there are significant differences in the prevalence of refractive errors among ethnic groups, even after controlling for age and sex (Kleinstein 2003). In a study of the US, Western European, and Australian year 2000 populations, the estimated crude prevalence for hyperopia of +3 dioptres (D) or greater in people aged 40 years or older was 9.9%, 11.6%, and 5.8%, respectively (Kempen 2004). 
 Hyperopic patients may be able to see well in the distance, provided they are young and the refractive error is not too high. Children with significant hypermetropia may complain of frequent headaches, rub their eyes often, have a lack of interest in or have difficulty reading. Young children with significant hypermetropia can also develop a convergent squint. Hypermetropia accompanied by significant anisometropia (where the two eyes have unequal refractive power) in children may lead to amblyopia if left untreated. 
 Uncorrected or undercorrected hyperopic adults may experience blurred vision ‐ especially when viewing near objects ‐ and asthenopia (visual discomfort, including burning eye sensation and tearing, fatigue after reading even for short periods and even bifrontal headaches exacerbated by near work) (Wilson II 1996).

Description of the intervention

Spectacles and contact lenses are the most commonly used methods to correct a refractive error. Both these methods are relatively safe, non‐invasive and reversible. There is a variety of surgical techniques available, aiming at modifying the lenticular refractive power, the corneal refractive power, or both. 
 The lenticular refractive power can be modified by phakic intraocular lens insertion; where the crystalline lens is maintained or by refractive lens exchange; where the crystalline lens is removed and replaced by an intraocular lens implant. 
 Several laser and non‐laser refractive surgical procedures have been used to modify the shape of the cornea and correct a refractive error (be it myopia, hyperopia and/or astigmatism), thereby restoring the focus plane of parallel light on the retina. The safety, efficacy, and predictability of the surgical outcomes have greatly improved since the introduction of the excimer laser and continue to improve as more advanced ophthalmological instruments become available. Despite these advances however, certain limitations and complications (infection, ectasia, diffuse lamellar keratitis, subepithelial haze, dry eye, epithelial ingrowth, buttonholed flap, free‐cap etc.) still exist (Sakimoto 2006).

How the intervention might work

In hyperopia, both photorefractive keratectomy (PRK) and laser assisted In situ keratomileusis (LASIK) employ the ophthalmic excimer laser in order to steepen the cornea and thus increase its refractive power. In the case of PRK the excimer laser is applied on the surface of the cornea after the corneal epithelium has been removed. In LASIK the creation of a thin corneal flap (with the aid of a blade or a laser beam) is followed by the excimer laser application on the underlying corneal tissue.

Why it is important to do this review

It is accepted that PRK is associated with more postoperative pain than LASIK, a slower visual recovery and a higher incidence of corneal haze (El‐Agha 2000; El‐Agha 2003a; El‐Agha 2003b). To our knowledge the question of which of these two methods for correction of hyperopia is more accurate, more stable and more safe than the other has not been addressed to date by a systematic review. The purpose of this review was to evaluate data from randomised controlled trials (RCTs) in order to address this question.

Objectives

The objective of this review was to compare PRK to LASIK for the correction of hyperopia by evaluating the postoperative uncorrected visual acuity, the refractive outcome, the potential loss of best spectacle corrected visual acuity, the postoperative pain and the incidence of adverse outcomes such as subepithelial haze, flap‐related complications and corneal ectasia.

Methods

Criteria for considering studies for this review

Types of studies

As both procedures have evolved and their outcomes improved in the last few years due to the advent of wavefront and corneal topographic analysis software, as well as faster and more efficient excimer laser delivery systems with iris registration and tracking technologies, it is important to compare the latest techniques for PRK with the latest techniques for LASIK. We therefore aimed to include all RCTs in this review but would then perform a sensitivity analysis of pre‐ and post‐millennial trials since this is the midpoint in the history of PRK and LASIK.

Types of participants

We only considered trials in which the participants were males or females over 18 years of age and underwent PRK or LASIK for correction of any degree of primary hyperopia, including hyperopic astigmatism. It has been postulated that excimer laser correction of hyperopia and hyperopic astigmatism in children and adolescents has the potential to correct refractive errors, improve visual function in amblyopic eyes, correct accommodative strabismus, and improve binocular vision (Dvali 2005). However, as the refractive error of patients under the age of 18 tends to change as part of their somatic development, we decided to exclude studies with such participants from our review. 
 Patients with a history of previous refractive or other eye surgery, co‐existing ocular pathology or systemic disease that affects wound healing were also excluded.

Types of interventions

We included studies in which PRK was compared to LASIK for correction of any degree of hyperopia, including hyperopic astigmatism.

Types of outcome measures

Primary outcomes

Our primary outcome was the percentage of eyes within ±0.50 D of target refraction at 12 months post‐treatment.

Secondary outcomes

Our secondary outcomes were as follows. 
 i. Percentage of eyes within ±0.50 D of target refraction at six months or less post‐treatment. 
 ii. Percentage of eyes that lost 2 or more lines of best spectacle corrected visual acuity (BSCVA) at 12 months post‐treatment. 
 iii. Percentage of eyes that lost 2 or more lines of best spectacle corrected visual acuity (BSCVA) at six months or less post‐treatment. 
 iv. Percentage of eyes with uncorrected visual acuity (UCVA) of 20/20 or better at less than six months or less post‐treatment. 
 v. Percentage of eyes with uncorrected visual acuity (UCVA) of 20/20 or better at 12 months post‐treatment. 
 vi. Percentage of eyes with uncorrected visual acuity (UCVA) of 20/40 or better at six months or less post‐treatment. 
 vii. Percentage of eyes with uncorrected visual acuity (UCVA) of 20/40 or better at 12 months post‐treatment.

Adverse outcomes

i. Incidence of corneal ectasia. 
 ii. Incidence of flap‐associated complications in eyes that underwent LASIK. 
 iii. Incidence of sub‐epithelial haze in eyes that underwent PRK. 
 iv. Postoperative pain scores. 
 v. Incidence of symptoms attributed to dry eyes post‐treatment.

Search methods for identification of studies

Electronic searches

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) 2012, Issue 2, part of The Cochrane Library. www.thecochranelibrary.com (accessed 17 February 2012), MEDLINE (January 1950 to February 2012), EMBASE (January 1980 to February 2012), Latin American and Caribbean Literature on Health Sciences (LILACS) (January 1982 to February 2012), the metaRegister of Controlled Trials (mRCT) (www.controlled‐trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the WHO International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). There were no language or date restrictions in the search for trials. The electronic databases were last searched on 17 February 2012.

See: Appendices for details of search strategies for CENTRAL (Appendix 1), MEDLINE (Appendix 2), EMBASE (Appendix 3), LILACS (Appendix 4), mRCT (Appendix 5), ClinicalTrials.gov (Appendix 6) and the ICTRP (Appendix 7).

Searching other resources

When trials are included in the review, we will search the reference lists of the studies included in the review for information about further trials. We will use the Science Citation Index to search for papers that cite any studies included in this review. We did not handsearch journals or conference proceedings specifically for this review.

Data collection and analysis

Selection of studies

All three review authors working independently assessed the titles and abstracts resulting from the searches. The full copies of all relevant studies were screened by all three authors against the inclusion criteria.

No relevant RCTs were identified from our search, but if we do find any RCTs in the future, we will follow the methodological process below.

Data extraction and management

Two review authors (SG and SC) will extract data independently using a standard data collection form and will enter data independently into Review Manager 5 (Review Manager 2011). The results will be compared and any disagreements will be resolved by discussion. Details will be extracted from the studies on the following: methods, participants, interventions, outcomes and notes.

Assessment of risk of bias in included studies

All three review authors will assess studies that meet the inclusion criteria for methodological quality, as described in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We will consider the following parameters: 
 i. random sequence generation; 
 ii. allocation concealment; 
 iii. masking (blinding) of participants, personnel and outcome assessors; 
 iv. incomplete outcome data; 
 v. selective outcome reporting; 
 vi. other sources of bias.

Each review author will grade each parameter independently as low risk of bias, high risk of bias or unclear (uncertain risk of bias). Studies that score unclear in any of the parameters will be dealt with by contacting the authors for clarification and additional information. Potential disagreements between the three review authors will be resolved by discussion. We will include all trials in our analysis initially, except those graded as high risk of bias on random sequence generation.

Measures of treatment effect

All outcome measures stated are dichotomous, with the exception of 'postoperative pain scores' which is ordinal. For dichotomous outcomes an odds ratio will be calculated. For continuous outcomes, the mean difference will be calculated.

Unit of analysis issues

The preferred unit of analysis will be outcomes for eyes rather than individuals, since some individuals might have had unilateral treatment or different treatment in each eye.

Dealing with missing data

In case we are unable to extract all the information we are interested in from published reports, both with regard to the details of the study and its numerical results, we will request the missing data from the original investigators.

Assessment of heterogeneity

When we review the studies, we will endeavour to identify differences between them which are likely to introduce heterogeneity. As some degree of heterogeneity will always exist due to the clinical and methodological diversity of the studies, we will employ the results of the I² measure to quantify inconsistencies across studies.

Assessment of reporting biases

In order to investigate whether our review is subject to reporting biases, we will use RevMan to prepare the relevant funnel plot which will be examined for signs of asymmetry as detailed in Chapter 10 of the Cochrane Handbook for Systematic Reviews of Interventions (Sterne 2011).

Data synthesis

If there is no substantial statistical heterogeneity, and if there is no clinical heterogeneity between the trials, we will combine the results in a meta‐analysis using a random‐effects model. A fixed‐effect model will be used if the number of trials is three or less. If heterogeneity is present, we will not combine the studies in a meta‐analysis but will present the details in a tabulated summary.

Subgroup analysis and investigation of heterogeneity

We will initially analyze data from all eligible studies. We will then perform subgroup analyses to investigate for heterogeneity for the following subgroups of hyperopia: low to moderate hyperopia (≤ +4.0 D) and moderate to high hyperopia (> +4.0 D).

Sensitivity analysis

Sensitivity analyses will be performed, in order to evaluate how robust the results of the review are relative to decisions and assumptions made in the process of conducting the review. We will include all trials in our analysis initially, except those graded as 'high risk of bias' on 'sequence generation'. We will then examine the effect of excluding trials graded as 'high risk of bias' on any of the parameters by repeating the analysis without these.

Results

Description of studies

Results of the search

The electronic searches yielded 339 titles and abstracts. The Trials Search Co‐ordinator (TSC) scanned the search results and removed any references which were not relevant to the scope of the review. We obtained the full text copies of five studies, which after further assessment were all excluded as none were RCTs.

An update search was done in February 2012 which yielded a further 12 references. The TSC scanned the search results and removed 11 references which were not relevant to the scope of the review. We screened the remaining reference which is an ongoing study from ClinicalTrials.gov. Currently there is not enough information to adequately assess this study for potential inclusion in the review therefore we will assess this study when further information becomes available.

Included studies

No studies met the inclusion criteria.

Excluded studies

We excluded five studies (El‐Agha 2000; El‐Agha 2003a; El‐Agha 2003b; Sciscio 2003; Stein 1999). For further details see the 'Characteristics of excluded studies' table.

Risk of bias in included studies

As no studies met the inclusion criteria, risk of bias was not assessed.

Effects of interventions

As no studies met the inclusion criteria, no meta‐analysis was performed.

Discussion

Our search identified no RCTs. We therefore discuss the data available from the five excluded studies comparing PRK against LASIK for correction of hyperopia and hyperopic astigmatism. These studies were non‐randomised and both 'sequence generation' and 'allocation concealment' were open to bias.

In a large, prospective, non‐randomised study by Stein et al. 200 eyes underwent PRK and 186 LASIK with the VISX STAR Excimer Laser System for hyperopia up to +8.00 D and followed up to 24 months (Stein 1999). This study examined the postoperative spherical equivalent (SE), uncorrected visual acuity (UCVA), best spectacle corrected visual acuity (BSCVA), epithelial healing time and complication rate. It was demonstrated that although similar SE was achieved in the long term, eyes that underwent LASIK experienced less discomfort, faster visual recovery and less hyperopic regression. The incidence of significant loss of BSCVA (2 lines) was 1% in both groups, while the complication rate was 4.5% in the PRK group and 1.6% in the LASIK group. The epithelial healing time was predictably longer for the PRK group.

These results are in keeping with the findings of later studies which compared PRK against LASIK for the correction of spherical hyperopia (El‐Agha 2000; El‐Agha 2003a) and compound hyperopic astigmatism (El‐Agha 2003b). In all three studies PRK and LASIK were deemed of comparable efficacy and safety. PRK was associated with significant postoperative pain, initial and temporary myopic overcorrection which self‐corrected within three to six months and an incidence of mild peripheral haze up to 19.5%. LASIK was less painful and was associated with more rapid refraction stability (at one month) and less initial myopic overcorrection. The long term efficacy and stability of astigmatic correction were similar in both groups.

A separate study utilised Fourier analysis of corneal topographic data to determine differences in induced irregular corneal astigmatism following spherical hyperopic correction by PRK versus LASIK (Sciscio 2003). It concluded that both procedures induced an equal amount of irregular corneal astigmatism which peaked at three months post‐op and then decreased over the next nine months, with no statistically significant difference between the PRK and LASIK group at any time point.

Authors' conclusions

Implications for practice.

To date there are only a few studies comparing PRK with LASIK directly for hyperopia correction. These studies are non‐randomised and as a rule include a small number of patients. Despite general agreement regarding the efficacy, safety and stability of postoperative refraction of hyperopic PRK or LASIK, firm and robust conclusions cannot be deduced from the available data.

Implications for research.

Well‐planned, large, high quality RCTs comparing hyperopic‐PRK with hyperopic‐LASIK are required, in order to provide a robust and reliable clinical evidence base. These studies must have a random sequence generation for the participants and the allocation sequence must be adequately concealed. As masking of the surgeon or the patient to the administered surgical intervention is nigh impossible, it is unavoidable that the trials will be open. The assessor of the primary and secondary outcomes however could and should be masked to the surgical intervention.

What's new

Date	Event	Description
3 May 2012	New search has been performed	Issue 6, 2012: Updated searches yielded one potential ongoing trial ongoing NCT01135719.
3 May 2012	New citation required but conclusions have not changed	Issue 6, 2012: Search sections updated and risk of bias judgements amended in this version.

History

Protocol first published: Issue 2, 2008
 Review first published: Issue 2, 2009

Date	Event	Description
26 March 2008	Amended	Converted to new review format.

Appendices

Appendix 1. CENTRAL search strategy

#1 MeSH descriptor Hyperopia 
 #2 hyperop* or hypermetrop* 
 #3 far next sight* 
 #4 long next sight* 
 #5 farsight* 
 #6 longsight* 
 #7 (#1 OR #2 OR #3 OR #4 OR #5 OR #6) 
 #8 MeSH descriptor Photorefractive Keratectomy 
 #9 keratectom* 
 #10 PRK 
 #11 (#8 OR #9 OR #10) 
 #12 MeSH descriptor Keratomileusis, Laser In Situ 
 #13 keratomileusis 
 #14 LASIK 
 #15 (#12 OR #13 OR #14) 
 #16 (#7 AND #11 AND #13)

Appendix 2. MEDLINE (OvidSP) search strategy

1. randomized controlled trial.pt. 
 2. (randomized or randomised).ab,ti. 
 3. placebo.ab,ti. 
 4. dt.fs. 
 5. randomly.ab,ti. 
 6. trial.ab,ti. 
 7. groups.ab,ti. 
 8. or/1‐7 
 9. exp animals/ 
 10. exp humans/ 
 11. 9 not (9 and 10) 
 12. 8 not 11 
 13. hyperopia/ 
 14. (hyperop$ or hypermetrop$).tw. 
 15. (far adj1 sight$).tw. 
 16. (long adj1 sight$).tw. 
 17. farsight$.tw. 
 18. longsight$.tw. 
 19. or/13‐18 
 20. photorefractive keratectomy/ 
 21. keratectom$.tw. 
 22. PRK.tw. 
 23. or/20‐22 
 24. keratomileusis, laser in situ/ 
 25. keratomileusis.tw. 
 26. LASIK.tw. 
 27. or/24‐26 
 28. 19 and 23 and 27 
 29. 12 and 28

The search filter for trials at the beginning of the MEDLINE strategy is from the published paper by Glanville et al (Glanville 2006).

Appendix 3. EMBASE (OvidSP) search strategy

1. exp randomized controlled trial/ 
 2. exp randomization/ 
 3. exp double blind procedure/ 
 4. exp single blind procedure/ 
 5. random$.tw. 
 6. or/1‐5 
 7. (animal or animal experiment).sh. 
 8. human.sh. 
 9. 7 and 8 
 10. 7 not 9 
 11. 6 not 10 
 12. exp clinical trial/ 
 13. (clin$ adj3 trial$).tw. 
 14. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$)).tw. 
 15. exp placebo/ 
 16. placebo$.tw. 
 17. random$.tw. 
 18. exp experimental design/ 
 19. exp crossover procedure/ 
 20. exp control group/ 
 21. exp latin square design/ 
 22. or/12‐21 
 23. 22 not 10 
 24. 23 not 11 
 25. exp comparative study/ 
 26. exp evaluation/ 
 27. exp prospective study/ 
 28. (control$ or prospectiv$ or volunteer$).tw. 
 29. or/25‐28 
 30. 29 not 10 
 31. 30 not (11 or 23) 
 32. 11 or 24 or 31 
 33. hypermetropia/ 
 34. (hyperop$ or hypermetrop$).tw. 
 35. (far adj1 sight$).tw. 
 36. (long adj1 sight$).tw. 
 37. farsight$.tw. 
 38. longsight$.tw. 
 39. or/33‐38 
 40. photorefractive keratectomy/ 
 41. keratectom$.tw. 
 42. PRK.tw. 
 43. or/40‐42 
 44. keratomileusis/ 
 45. keratomileusis.tw. 
 46. LASIK.tw. 
 47. or/44‐46 
 48. 39 and 43 and 47 
 49. 32 and 48

Appendix 4. LILACS search strategy

hyperopia or hypermetropia and keratectom$ or PRK and keratomileusis or LASIK

Appendix 5. metaRegister of Controlled Trials search strategy

hyperopia and keratectomy and keratomileusis

Appendix 6. ClinicalTrials.gov search strategy

Hyperopia AND Keratectomy AND Keratomileusis

Appendix 7. ICTRP search strategy

Hyperopia AND Keratectomy AND Keratomileusis

Characteristics of studies

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
El‐Agha 2000	Non‐randomised study
El‐Agha 2003a	Non‐randomised study
El‐Agha 2003b	Non‐randomised study
Sciscio 2003	Non‐randomised study
Stein 1999	Non‐randomised study

Characteristics of ongoing studies [ordered by study ID]

NCT01135719.

Trial name or title	A comparison of fellow eyes undergoing LASIK or PRK with a wavefront‐guided excimer laser versus a wavefront‐optimised excimer laser
Methods	Subjects with myopia, hyperopia and astigmatism are being randomised by ocular dominance to be treated with either PRK or LASIK. One eye will be treated with a wavefront‐guided excimer laser and the fellow eye will be treated with a wavefront optimised excimer laser
Participants	Participants 21 years and older
Interventions	Wavefront‐guided excimer laser versus wavefront‐optimised excimer laser
Outcomes	Changes in best spectacle corrected visual acuity Changes in 25 and 5% low contrast visual acuity Improvement in uncorrected visual acuity Changes in higher order aberrations
Starting date	April 2009
Contact information	Edward E. Manche, Stanford University School of Medicine, Stanford University
Notes

Differences between protocol and review

Due to an error in the protocol, the primary outcome stated in the review is different. The primary outcome stated in the published protocol was percentage of eyes with uncorrected visual acuity (UCVA) of 20/20 or better at 12 months post‐treatment.

Contributions of authors

Conceiving the review: SG 
 Designing the review: SG, SC 
 Co‐ordinating the review: SG 
 Data collection for the review 
 ‐ Designing electronic search strategies: Cochrane Eyes and Vision Group Trials Search Co‐ordinator 
 ‐ Designing other search strategies: SG, SC, ME 
 ‐ Undertaking searches: SG, SC 
 ‐ Screening search results: SG, SC, ME, IY 
 ‐ Translation of Chinese papers: IY 
 ‐ Organising retrieval of papers: SG 
 ‐ Screening retrieved papers against inclusion criteria: SG, SC, ME 
 ‐ Appraising quality of papers: SG, SC, ME 
 ‐ Extracting data from papers: SG, SC 
 ‐ Writing to authors of papers for additional information: SG, SC 
 ‐ Providing additional data about papers: SG, SC 
 ‐ Obtaining and screening data on unpublished studies: SG, SC 
 Data management for the review 
 ‐ Entering data into RevMan: SG, SC 
 Analysis of data: SG, IY 
 Interpretation of data 
 ‐ Providing a methodological perspective: SG, SC 
 ‐ Providing a clinical perspective: SG, ME 
 ‐ Providing a policy perspective: SG, SC 
 ‐ Providing a consumer perspective: SG, SC 
 Writing the review: SG, SC, ME 
 Providing general advice on the review: SG 
 Securing funding for the review: SG 
 Performing previous work that was the foundation of the current study: SG, ME

Declarations of interest

None known.

New search for studies and content updated (no change to conclusions)