Article Text
PDF
Abstract
Background Despite extensive knowledge on the functional, neurodevelopmental, behavioural and educational sequelae of preterm birth, relatively little is known about its economic consequences.
Objective To systematically review evidence around the economic consequences of preterm birth for the health services, for other sectors of the economy, for families and carers, and more broadly for society.
Methods Updating previous reviews, systematic searches of Medline, EconLit, Web of Science, the Cochrane Library, Cumulative Index to Nursing and Allied Health Literature, Embase and Scopus were performed using broad search terms, covering the literature from 1 January 2009 to 28 June 2017. Studies reporting economic consequences, published in the English language and conducted in a developed country were included. Economic consequences are presented in a descriptive manner according to study time horizon, cost category and differential denominators (live births or survivors).
Results Of 4384 unique articles retrieved, 43 articles met the inclusion criteria. Of these, 27 reported resource use or cost estimates associated with the initial period of hospitalisation, while 26 reported resource use or costs incurred following the initial hospital discharge, 10 of which also reported resource use or costs associated with the initial period of hospitalisation. Only two studies reported resource use or costs incurred throughout the childhood years. Initial hospitalisation costs varied between $576 972 (range $111 152–$576 972) per infant born at 24 weeks’ gestation and $930 (range $930–$7114) per infant born at term (US$, 2015 prices). The review also revealed a consistent inverse association between gestational age at birth and economic costs regardless of date of publication, country of publication, underpinning study design, follow-up period, age of assessment or costing approach, and a paucity of evidence on non-healthcare costs. Several categories of economic costs, such as additional costs borne by families as a result of modifications to their everyday activities, are largely overlooked by this body of literature. Moreover, the number and coverage of economic assessments have not increased in comparison with previous review periods.
Conclusion Evidence identified by this review can be used to inform clinical and budgetary service planning and act as data inputs into future economic evaluations of preventive or treatment interventions. Future research should focus particularly on valuing the economic consequences of preterm birth in adulthood.
- preterm birth
- health economics
- costing
Statistics from Altmetric.com
What is already known on this topic?
There is extensive knowledge on the functional, neurodevelopmental, behavioural and educational sequelae of preterm birth.
Evidence on the economic consequences of preterm birth is more limited, with previous reviews of the topic extending only to 2009.
What this study adds?
This systematic review reveals no evidence that the number or coverage of economic assessments has increased.
The review reveals a consistent inverse association between gestational age at birth and economic costs; and a paucity of evidence on non-healthcare costs.
Evidence identified by this review can be used to inform clinical and budgetary service planning and act as data inputs into future economic evaluations.
Introduction
Preterm birth has been defined by the WHO as any birth before 37 completed weeks’ gestation, or fewer than 259 days since the first day of the mother’s last menstrual period.1 It has been further subdivided into subcategories based on gestational age at birth, including extremely preterm (<28 weeks’ gestation), very preterm (28 to <32 weeks’ gestation), moderately preterm (32 to <34 weeks’ gestation) and late preterm (34 to <37 weeks’ gestation).2 Almost 15 million babies worldwide were born preterm in 2010, representing 11.1% of all live births, with prevalence ranging from approximately 5% in several northern European countries to 18% in Malawi.2 Moreover, time trends based on data from 65 countries with reliable time trend data between 1990 and 2010 and more than 10 000 live births per year reveal that the preterm birth rate has either remained stable (n=14) or increased (n=48) in 62 of those 65 countries.2
Mortality as a result of preterm birth complications ranks, together with pneumonia, as the leading cause of childhood deaths at the global level with over three million related annual neonatal deaths worldwide.3 Although survival rates have improved in recent years, this masks considerable variation in the chances of survival across4 and within5 countries. Furthermore, surviving children are at higher risk of cerebral palsy, visual and auditory deficits, poor respiratory outcomes, impaired motor and cognitive ability, and psychiatric disorders than children born at term.6–9 Up to one-third of children born extremely or very preterm and their parents face a life course with significant morbidity, dependency and socioeconomic challenges.10
Despite an extensive body of knowledge on the functional, neurodevelopmental, behavioural and educational sequelae of preterm birth,11 12 relatively little is known about its economic consequences. This paper presents a systematic review of the recent scientific literature on the economic consequences of preterm birth for the health services, for other sectors of the economy, for families and carers, and more broadly for society. The paper represents an update of previous review articles of the topic published by one of the authors (SP) that had covered the periods 1980–1999 and January 2000–June 2009, respectively.13 14
Methods
This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.15 A comprehensive literature search strategy was developed and piloted, building on our previous reviews of the topic.13 14 The final search strategy encompassed granulated thesauri terms and free-text search terms associated with preterm birth (eg, preterm or premature* or low birthweight or gestational age) intersected with terms associated with economic consequences (eg, ‘cost*’ or ‘economic burden*’ or ‘financial burden*’ or ‘resource expenditure*’ or ‘health care cost*’ or ‘cost of illness’), and is presented in full in online supplementary appendix 1. The time horizon of the search strategy covered the period between 1 January 2009 and 28 June 2017. The electronic databases searched included Medline, EconLit, Web of Science (including Index to Scientific & Technical Proceedings, Science Citation Index, Social Science Citation Index), the Cochrane Library (including York Database of Abstracts of Reviews of Effectiveness, NHS Economic Evaluation Database), Cumulative Index to Nursing and Allied Health Literature, Embase, and Scopus. Our search strategy was supplemented by manual reference searching of bibliographies, contacts with experts in the field, citation searching and author searching. All articles identified by the searches were entered into EndNote V.7.7 and duplicates were removed.
Supplemental material
The main inclusion criteria framing the search strategy were primary studies reporting economic consequences associated with preterm birth (<37 weeks’ gestation) or low birth weight (<2500 g), published in the English language and conducted in a developed country (defined, for the purposes of this review, as a member of the Organisation for Economic Co-operation and Development (OECD)). A two-stage screening process was followed. Titles and abstracts were assessed at the first stage of the review by two independent reviewers (HHY and JK). If an article received approval from both reviewers, it proceeded to the next stage, with disagreements referred to a third reviewer (SP) for final assessment. At the second stage, two reviewers (HHY and SP) independently reviewed full-text articles to assess whether they met the review’s inclusion criteria, with disagreements resolved via consensus. The term ‘article’ is used in its broadest sense and encompasses monographs, book chapters and conference abstracts. Studies of the economic consequences of low birth weight were included at the second stage of the review in order to capture potentially relevant information from contexts of clinical uncertainty in and incomplete recording of estimates of gestation. In addition, studies that reported the resource consequences associated with preterm birth or low birth weight in metrics amenable to future economic analyses were included at the second stage of the review for completeness. We excluded studies at the full article stage that (1) were not relevant to the economic aspects of preterm birth or low birth weight; (2) did not report primary research evidence; (3) were not published in the English language; (4) treated preterm birth or low birth weight as a covariate rather than as a dependent variable or outcome; or (5) provided insufficient information to arrive at an economic assessment.
From each article that met the study selection criteria, we extracted the following information about the characteristics of the study using a bespoke proforma and entered it into an Excel database: (1) bibliographic details, including year of publication; (2) date of cohort/study; (3) country/geographical jurisdiction; (4) study design; (5) sample size(s); (6) gestational age(s) at birth of study participants; (7) birth weight(s) of study participants; (8) period over which study participants were followed up; (9) cost categories; (10) resource use categories; (11) currency in which costs were expressed; (12) financial year for which costs were valued; and (13) cost data sources for the valuation process. The quality of contributing studies was assessed using a subset of 18 relevant items selected from the 24-item Consolidated Health Economic Evaluation Reporting Standards checklist for health economic evaluations.16 Cost data extracted from studies were inflated, where necessary, to 2015 prices using the relevant country-specific gross domestic product deflator index, and subsequently converted, where necessary, from their respective currencies into US dollars using purchasing power parities supplied by the OECD.17 For studies that failed to report their currency price dates, it was assumed that the costs used in the valuation process applied to the financial year prior to the publication of the study.
Methodological variations between studies, including variations in underpinning healthcare practices across jurisdictions and variations in the relative prices of labour and capital inputs across jurisdictions, prevented a pooling of economic data akin to the meta-analyses performed on clinical effectiveness estimates. Rather, resource use and economic cost estimates are presented in a descriptive manner according to study time horizon, resource use and/or cost category, and differential denominators (live births or survivors).
Results
A total of 4384 unique articles were identified by the combined literature searches (figure 1). Of these, only 107 articles satisfied the first stage of inclusion criteria, while a further 64 articles were excluded at the second stage of the review process. A total of 43 articles met the inclusion criteria at both stages and were included in this systematic review.18–60 Of these 43 articles, 27 reported resource use or cost estimates associated with the initial period of hospitalisation,18–21 23–26 29 31 36–38 40 42–44 46–49 51 53 54 56 57 59 while 26 reported resource use or costs incurred following the initial hospital discharge,21–23 25 27 28 30 32–36 38–41 45–47 50 52 54–56 58 60 10 of which also reported resource use or costs associated with the initial period of hospitalisation.21 23 25 36 38 40 46 47 54 56 Only two studies reported resource use or costs incurred throughout the childhood years.33 46 The quality of studies that met the inclusion criteria of the review is summarised in online supplementary appendix 2, with quality assessment scores ranging from 224 44 48 51 to 17.35 40 57 Notable methodological limitations included a failure to specify price dates for cost estimates,18 21 23 34 43 44 47 48 51 failure to report discount rates for longer terms costs22 30 33 34 50 54 55 58 and reliance on charges, which are likely to have included elements arising from corporate financial decisions.19
PRISMA flow diagram. PRISMA denotes Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
The methodological characteristics of studies reporting resource use and/or economic cost estimates associated with the initial hospitalisation, including their underpinning design, sample size, study population, study perspective and data sources, are summarised in brief format in table 1 and in full format in online supplementary appendix 3. The study estimates of hospital length of stay and economic costs are presented for differential denominators (live births or survivors) in brief format in table 2 and in full format in online supplementary appendix 4. Initial hospitalisation costs varied between $576 97229 (range $111 15257–$576 97229) per infant born at 24 weeks’ gestation and $93043 (range $93043–$711459) per infant born at term, and from $169 132 per surviving infant born at <1500 g51 to $1200 per infant born at ≥2500 g21 (US$, 2015 prices). The duration of the initial hospitalisation varied between a median of 11631 days (range of a mean of 8453 to a median of 11631 days) per surviving infant born at 24 weeks’ gestation and a mean of 2.421 days (range 2.421–5.242) per infant born at term. A consistent inverse association was observed between gestational age at birth and economic costs regardless of date of publication, country of publication, underpinning study design or costing approach. The only exceptions were three studies that showed that, among infants born at ≤26 weeks’ gestation, costs increased as gestational age increased.44 57 59 The denominators for these analyses were live births and therefore included infants that died as well as those that survived; the implication is that initial hospitalisation costs increased with gestational age in these cohorts as a consequence of the improved survival chances of the least immature. Only two studies estimated maternal costs to understand additional resources that may be associated with preterm birth.29 47 Both studies revealed an inverse association between gestational age at birth and mean maternal costs. Moreover, only one study estimated non-healthcare costs associated with the initial period of hospitalisation.25 In a prospective cohort study of 150 very low birthweight (VLBW) infants without prematurity-related morbidities and 145 full-term controls, Cavallo et al used parental questionnaires to estimate that travel costs borne by parents, and the value of lost productivity represented 7.2% and 29.6% of the total societal costs, respectively, in the VLBW group, and 1.6% and 64.1% of the total societal costs, respectively, in the control group.25 Another notable feature of economic studies that focused on the initial hospitalisation is the observed independent effect of multiplicity on increased hospitalisation costs.43 57
Studies reporting resource use and/or economic costs associated with the initial hospitalisation (full methodological characteristics presented in online supplementary appendix 3)
Summary of economic consequences associated with the initial hospitalisation (full results presented in online supplementary appendix 4)
The methodological characteristics of studies reporting resource use and/or economic costs following initial hospitalisation are summarised in online supplementary appendix 5, with study estimates of economic outcomes presented in online supplementary appendix 6. Ten studies estimated outcomes during the first 2 years of life,21 23 25 28 36 41 45 47 56 60 with the remainder of the studies estimating economic outcomes over longer follow-up periods or at later ages within cross-sectional studies. A consistent inverse association was observed between gestational age at birth and economic costs regardless of period of follow-up and age at assessment. This pattern held when children born moderate or late preterm were compared with children born at term,22 23 27 30 36 41 47 55 60 and when children born at early term were compared with children born at late term.30 41 55 60 Longitudinal studies with assessments of economic outcomes at repeated time points revealed that health service resource use and costs declined as children aged, regardless of gestational age at birth.27 30 38 50 54 Eight studies estimated economic costs accruing postinitial discharge, associated with preterm birth, which are borne outside of the health sector.25 30 35 36 39 46 47 52 Non-health service costs exceeded health service costs in the studies by Cavallo et al,25 which was limited to an 18-month time horizon, and the study by Petrou et al,52 which was limited to the 11th year of life, although no study valued economic costs borne by all sectors of the economy and by caregivers.
The two studies that reported economic costs throughout the childhood years used different methodological approaches. The earlier study, by Mangham et al,46 applied decision-analytic modelling to estimate public sector costs associated with preterm birth. The model assumed a hypothetical cohort of children, equivalent in size to the number of live births in England and Wales in 2006, and was populated using data from thee cohort studies, namely EPICure, the Victorian Infant Collaborative Study and the Oxford Record Linkage Study. The mean incremental public sector cost per preterm child surviving to 18 years, compared with a term survivor, was estimated at $39 329 (US$, 2015 prices). The corresponding estimates for a very and extremely preterm child were substantially higher at $106 174 and $162 816, respectively. More recently, Hummer et al 33 applied econometric methods to Austrian health insurance administrative panel data linked to birth registry data. They found that although absolute differences in health service utilisation between children with birth weights <2500 g and ≥2500 g diminished over time, those born at low birth weight still spent significantly more days in hospital and generated significantly higher medical drug and medical assistance expenses through early childhood. Furthermore, there was a shift during the years of compulsory schooling towards care for diseases of the nervous system and mental and behavioural disorders among children born at low birth weight with effects persisting until early adulthood.
Discussion
This paper systematically reviews the recent (2009–2017) scientific literature on the economic consequences of preterm birth for alternative stakeholders, updating two previous review articles of the topic.13 14 The number of relevant studies identified by this review was smaller than the number of studies covered by the most recent review (43 vs 52), which covered a comparable time horizon, although the number of countries in which studies were conducted increased (13 vs 6).14 In addition, a smaller proportion of studies estimated economic consequences following the initial hospital discharge (60.5% vs 67.3%) and economic consequences outside the health sector (18.6% vs 25.0%).14 There does not appear to be evidence that the increased longevity of cohort studies of children born preterm or the establishment of collaborative research platforms around preterm birth (eg, http://www.apic-preterm.org/) has increased the number or coverage of economic assessments.
When viewed in conjunction with evidence from previous systematic reviews,13 14 the economic costs to the health services, other sectors of the economy, and to families and caregivers associated with preterm birth remain considerable. Comparisons of cost estimates with those reported for other childhood conditions are constrained by a number of methodological factors, including differences in costing methodologies, ages at assessment, study perspectives and periods of follow-up, as well as variations in underpinning healthcare practices and relative prices of resource inputs across study settings. Nevertheless, a holistic overview of the identified evidence suggests that ongoing economic costs associated with extremely preterm and very preterm birth exceed the annual cost burdens reported for several chronic childhood conditions, including asthma,61 juvenile idiopathic arthritis,62 depression,63 separation anxiety64 and attention-deficit hyperactivity disorder.65
Several categories of economic costs have, with a few exceptions,36 52 been overlooked by this body of literature. These include costs borne by local authorities and voluntary organisations, such as adaptations that have to be made to the individual’s home as a result of their impaired state of health, and additional costs borne by families as a result of modifications to their everyday activities. In addition to the costs of travel, childcare and accommodation, other potential costs faced by families and informal carers that have been largely overlooked include incremental expenditures on health goods, such as alternative therapies, and non-health goods, such as nutritional requirements, laundry, clothing, heating utilities and repairs to the home. Methods for ascertaining these costs, including questionnaires, diaries and data extraction from administrative databases, are available and widely applied in other areas of healthcare.66 The value of ascertaining these costs should be balanced against the additional burden it may impose on study participants.
The full effects of impairment associated with preterm birth on economic outcomes clarify only with time. Cohort studies with follow-up into adulthood and whole-country record linkage studies provide potential vehicles for ascertaining long-term economic outcomes such as long-term use of health and social care services, employment and occupational status, income, receipt of social welfare benefits and reproductive health, which might in turn have economic sequelae. As yet, no study has valued these long-term economic sequelae in adulthood. Recently established collaborative research platforms with a focus on outcomes in adulthood, such as the ‘Adults Born Preterm International Collaboration’ (http://www.apic-preterm.org/) and the ‘Research on European Children and Adults born Preterm’ programme (https://recap-preterm.eu/), should provide opportunities for economic research.
The strengths of this study are that it was conducted according to internationally agreed design and reporting guidelines for systematic reviews of economic evidence,67 and was broadly consistent in its methodology with previous reviews of the topic, thereby allowing intertemporal comparisons to be made.13 14 Limitations include exclusion of health utility-based measures of economic outcome that can be used to inform cost-effectiveness-based decision-making.68 It is likely that preterm birth is also associated with decrements in health utilities, that is, preference-based health-related quality of life outcomes,69 which have not been assessed in this review. Furthermore, the review did not extend to an assessment of the cost-effectiveness of preventive or treatment interventions for preterm birth.
The data identified by this review can be used to inform clinical and budgetary service planning and provide policy-relevant information for cross-country, longitudinal and other cost comparisons.70 The data can also act as inputs within cost-effectiveness models for preventive or treatment interventions. There are several circumstances, particularly in the context of decision-analytic modelling-based economic evaluations, where economic analysts lack the time and resources to collect primary data for economic parameters of interest. This review has generated a body of data that can act as inputs into future economic models, and allows the reader to assess the suitability of the evidence for their particular context.
In conclusion, this paper presents a systematic review of the recent scientific literature on the economic consequences of preterm birth for the health services, for other sectors of the economy, for families and carers, and more broadly for society. Future research should focus particularly on valuing the economic consequences of preterm birth in adulthood.
Acknowledgments
We would like to thank Samantha Johnson for her assistance with the development and piloting of the literature search strategies. We would also like to thank colleagues at the University of Warwick for comments on earlier drafts of the paper.
References
Footnotes
Contributors SP designed this study, reviewed the evidence from all contributing studies, wrote the article and acts as guarantor. HHY and JK screened all titles and abstracts. HHY also reviewed the evidence from all contributing studies.
Funding The Warwick Clinical Trials Unit, University of Warwick, benefited from facilities funded through the Birmingham Science City Translational Medicine Clinical Research and Infrastructure Trials Platform, with support from Advantage West Midlands. SP receives support as an NIHR Senior Investigator (NF-SI-0616-10103). SP also receives funding from the European Union’s 2020 research and innovation programme, RECAP, under grant agreement number 733280.
Disclaimer The views contained within this paper are those of the authors and not necessarily of the funders.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; externally peer reviewed.