Studies reporting resource use and/or economic costs associated with the initial hospitalisation (full methodological characteristics presented in online supplementary appendix 3)
| Study | Date of birth of study population | Location | Type of study | Sample size | Gestational age(s) (weeks) | Birth weight (g) |
| Allen et al 18 | 2012–2015 | USA | Case series | 23 weeks: 12 | 23–25 | <2500 |
| 24 weeks: 17 | ||||||
| 25 weeks: 22 | ||||||
| Almond et al 19 | 1983–2002 | USA | Retrospective analysis of national birth cohort | >200 000 newborns of 3 ounces around the 1500 g threshold in the nationwide data; approximately 30 000 births in the same interval for the five-state sample. | NA | 3 ounces (85 g) bandwidth from 1500 g threshold |
| State inpatient research database analysis | ||||||
| 1991–2006 | 5 US states (Arizona, California, Maryland, New Jersey and New York) | |||||
| Aly et al 20 | 2007–2008 | USA | Retrospective analysis of national birth cohort | All: 81 913 | 33–34 | NA |
| 33–34 weeks: 24 086 | 35–36 | |||||
| 35–36 weeks: 57 827 | ||||||
| Barradas et al 21 | 2009 | USA | Retrospective analysis of national birth cohort | 4 167 000 (9.1% were preterm or low birth weight) | <37 | <2500 |
| Bird et al 23 | 2001–2005 | USA (Arkansas) | Retrospective analysis of Medicaid claims database | 34–36 weeks: 5188 | 34–36 | NA |
| 37–42 weeks: 15 303 | 37–42 | |||||
| Bulut et al 24 | 2006–2008 | Turkey (Istanbul) | Retrospective chart analysis | 34–36 weeks: 6742 | 34–36 | NA |
| ≥37 weeks: 30 055 | ≥37 | |||||
| Cavallo et al 25 | 2006 | Italy | Prospective cohort study | ≤30 weeks and/or ≤1500 g: 150 | ≤30 | ≤1500 |
| 37–42 weeks: 145 | 37–42 | |||||
| Cömert et al 26 | 2006–2007 | Turkey (Istanbul) | Retrospective chart analysis | 28–32 weeks: 96 | 28–32 | NA |
| 33–37 weeks: 115 | 33–37 | |||||
| Hall and Greenberg29 | 2012 | USA (Hamilton County, Ohio) | Retrospective chart analysis | 10 976 births of which 1444 were preterm | 23–27 | NA |
| 28–31 | ||||||
| 32–36 | ||||||
| Hinchliffe et al 31 | 2006–2010 | UK (East Midlands and Yorkshire) | Population-based audit of neonatal care | 24 weeks: 296 | 24–28 | NA |
| 25 weeks: 384 | ||||||
| 26 weeks: 551 | ||||||
| 27 weeks: 663 | ||||||
| 28 weeks: 829 | ||||||
| Khan et al 36 | 2009–2010 | UK (East Midlands) | Prospective cohort study | 32–33 weeks: 162 | 32–33 | NA |
| 34–36 weeks: 984 | 34–36 | |||||
| 32–36 weeks: 1146 | 32–36 | |||||
| ≥37 weeks: 1258 | ≥37 | |||||
| Khashu et al 37 | 1999–2002 | Canada (British Columbia) | Retrospective analysis of regional birth cohort | 33–36 weeks: 6381 | 33–36 | NA |
| 37–40 weeks: 88 867 | 37–40 | |||||
| Korvenranta et al 38 | 2000–2003 | Finland | Retrospective national register study | 22–23 weeks: 17 | <32 | <1501 |
| 24–25 weeks: 135 | ||||||
| 26–27 weeks: 234 | ||||||
| 28–29 weeks: 391 | ||||||
| 30–31 weeks: 771 | ||||||
| >31 weeks: 248 | ||||||
| Korvenranta et al 40 | 2000–2003 | Finland | Retrospective national register study | Preterm survivors: | <32 | <1501 |
| 23 weeks: 17 | 37–41 | |||||
| 24–25 weeks: 135 | ||||||
| 26–27 weeks: 231 | ||||||
| 28–29 weeks: 385 | ||||||
| 30–31 weeks: 756 | ||||||
| Preterm non-survivors: 278 | ||||||
| Term: 200 609 | ||||||
| Leone et al 42 | 2006–2007 | Switzerland | Retrospective chart analysis | 34–36 weeks: 530 | 34–36 | NA |
| 39–40 weeks: 1686 | 39–40 | |||||
| Lim et al 43 | 2005–2006 | Canada | Retrospective national register study | Singletons: | <28 | NA |
| <28 weeks: 762 | 28–31 | |||||
| 28–31 weeks: 1099 | 32–33 | |||||
| 32–33 weeks: 1428 | 34–36 | |||||
| 34–36 weeks: 9716 | ≥37 | |||||
| ≥37 weeks: 189 750 | ||||||
| Multiples: | ||||||
| <28 weeks: 264 | ||||||
| 28–31 weeks: 418 | ||||||
| 32–33 weeks: 571 | ||||||
| 34–36 weeks: 2370 | ||||||
| ≥37 weeks: 2742 | ||||||
| Lo et al 44 | 2000–2008 | USA | Retrospective chart analysis | Singleton deliveries: 240 179 | 24–42 | NA |
| Mangham et al 46 | 2006 | England and Wales | Decision-analytic model populated with data from administrative population-based databases | ≤27 weeks: 3185 | All | NA |
| 28–31 weeks: 5575 | ||||||
| 32–33 weeks: 6410 | ||||||
| 34–36 weeks: 32 812 | ||||||
| ≥37 weeks: 621 618 | ||||||
| McLaurin et al 47 | 2004 | USA | Retrospective analysis of insurance database | 33–36 weeks: 1683 | 33–36 | NA |
| ≥37 weeks: 33 745 | ≥37 | |||||
| Meyers and Meyers48 | 2008 | USA | Retrospective analysis of national birth cohort | Not specified | All | All |
| Mistry et al 49 | 2001–2002 | UK (Basildon, Colchester and Southend in Essex and Gillingham in Kent) | Prospective cohort study | Three level 2 neonatal units: 192 | NA | ≤1800 |
| One level 3 neonatal unit: 199 | ||||||
| Nordermoen and Bratlid51 | 1997–2007 | Norway | Retrospective analysis of two birth cohorts | Not specified | <28 weeks ≥28 weeks | <1500 g |
| Seaton et al 53 | 2011–2014 | England | Retrospective analysis of electronic health records | 2011: 5368 | 24–31 | NA |
| 2012: 5343 | ||||||
| 2013: 5228 | ||||||
| 2014: 5099 | ||||||
| Stephens et al 54 | 2001–2011 | Australia (New South Wales) | Retrospective analysis of regional birth cohort | 24–27 weeks: 3659 | 24–27 | NA |
| 28–31 weeks: 7235 | 28–31 | |||||
| 32–33 weeks: 9007 | 32–33 | |||||
| 39–40 weeks: 539 631 | 39–40 | |||||
| Thanh et al 56 | 2004–2010 | Canada (Alberta) | Retrospective analysis of regional birth cohort | <2500 g: 16 209 | NA | <2500 g |
| ≥2500 g: 189 586 | ≥2500 g | |||||
| van Baaren et al 57 | 2006–2012 | The Netherlands | Retrospective analysis of one prospective cohort study and three randomised controlled trials | 4552 | 24–28 | NA |
| 1090 singletons | 28–32 | |||||
| 3462 multiples | 32–36 | |||||
| ≥37 | ||||||
| Xu et al 59 | 2003 | USA (Michigan) | Retrospective analysis of regional birth cohort | <37 weeks: 9780 | ≥20 | NA |
| ≥37 weeks: 101 484 |
NA, not applicable.