Skip to main content

Table 1 Key data fields that should be extracted and narratively synthesised by systematic reviews of falls prevention economic evaluations

From: Economic evaluation of community-based falls prevention interventions for older populations: a systematic methodological overview of systematic reviews

Category

Data field

(A) Setting, population and evaluation framework

1. Bibliography: author(s); publication year

2. Setting and aim: country; region; decision-maker; evaluation aim

3. Study design: e.g., decision model

4. Target population/sample demographics and comorbidities: e.g., residence – community-dwelling and/or institutionalised; age; sex; SES; health conditions unrelated to falls risk

5. Type of analysis: e.g., CUA, CEA, CBA, ROI

6. Perspective: e.g., public sector, societal

7. Cost-effectiveness threshold clearly stated

8. Time horizon of analysis/model

9. Discount rates (if time horizon is longer than 1 year)

(B) Falls epidemiology

1. Target population/sample falls risk factors/profile at baseline

2. Fall type: definition; recording method

3. Health consequences of falls: injury type; long-term consequences (e.g., institutionalisation, excess mortality risk)

4. Health utility measurement: acute vs. long-term impact of falls on health utility; comorbidity-related impact on health utility

5. Economic consequences of falls: care resource types; unit costs; all-cause and fall-related costsa

6. Wider/societal consequences of falls: e.g., social isolation from fear of falling; informal caregiver burden; productivity loss of older persons and caregivers

(C) Falls prevention intervention

1. Intervention characteristics: type (e.g., exercise, multifactorial); reach;b primary vs. secondary prevention; main components; staff type; duration, frequency and dose; mutual exclusivity;c comparator(s)

2. Intervention pathway: type (e.g., reactive, proactive, self-referredd); recruitment method; falls risk identification method; mutual exclusivity

3. Intervention resource use: e.g., staff labour and training; transport; overheads

4. Intervention costs: variable vs. fixed costs; economies of scale; societal costs (e.g., time opportunity cost, private co-payment)

5. Intervention implementation: uptake rate; adherence rate; sustainability rate

6. Intervention efficacy: risk of bias in estimation; match with incidence metric;e efficacy fall type;f efficacy durability;g wider health benefits; side effects

7. Intervention study characteristics: study design (e.g., RCT, meta-analysis); population/sample characteristicsh

(D) Decision model features

1. Model type and justification of type

2. Model cycle length and justification of length

3. Methods for adopting a long-term model horizoni

4. Methods for characterising baseline demographics and falls risk of model target population

5. Methods for characterising multiple falls in a year (recurrent falls)

6. Methods for characterising dynamic progression of falls risk factors, long-term consequences of falls and falls prevention intervention needj

7. Methods for characterising dynamic progression in comorbidities and changes in care costs, mortality risks, institutionalisation risks and health utilities

8. Methods for incorporating psychological and sociological variables (e.g., motives for healthy behaviour, community institutions) as determinants of falls risk, falls prevention access and model outcomes

9. Methods for incorporating budget and capacity constraints

10. Methods for reducing structural uncertainty of model prospectivelyk

11. Model validation methods/results: face; internal; external

(E) Evaluation methods and results

1. Cost-per-unit ratios (e.g., incremental cost per QALY gain)

2. Aggregate health and cost outcomes (e.g., total intervention cost, total QALY gain, total number of falls prevented)

3. Currency: original type/year; conversion to same currency for comparison

4. Handling heterogeneity: subgroup analyses; targeting analyses (under budget or capacity constraint)

5. Handling parameter uncertainty: deterministic sensitivity analysis; probabilistic sensitivity analysis

6. Scenario analyses: testing structural assumptions; scenario suggestions by stakeholders/decision-maker; value of implementation analysis [26]

7. Equity analyses: intervention impact on social inequities in health; estimating efficiency cost or joint equity-efficiency impact of prioritising vulnerable groups (e.g., via distributional cost-effectiveness analysis (DCEA) [27])

8. Model cross-validity: comparison of results to previous models

(F) Discussions by evaluation authors

1. Discussion on issues of generalisability and policy implementation

2. Discussion on strengths and limitations of evaluation

  1. Abbreviation: CBA cost–benefit analysis; CEA cost-effectiveness analysis, CUA cost-utility analysis, QALY quality-adjusted life year, RCT randomised controlled trial, ROI return on investment, SES socioeconomic status
  2. aExpert guideline on falls prevention economic evaluation recommends that evaluations report all-cause/total healthcare costs in the base case and fall-related costs in sensitivity analysis [22]
  3. bIntervention reach refers to the number/proportion of persons in the target population accessing the intervention. It is a function of intervention’s normative reach defined by its eligibility criteria and its implementation reach determined by implementation level (e.g., uptake rates) within the eligible population
  4. cSeveral intervention types and pathways can be non-mutually exclusive in a setting: e.g., reactive home assessment and modification for fallers discharged from hospitals and self-referred exercise
  5. dReactive pathway is accessed immediately after a fall requiring medical attention. Proactive pathway is accessed via referrals by care professionals in the community. Self-referred pathway is accessed voluntarily by older persons based on community/peer marketing
  6. eThis only concerns decision models that import falls efficacy evidence from external intervention studies. Main falls incidence metrics are falls risk and falls rate, and their matching efficacy metrics are relative risk (RR) and rate ratio (RaR), respectively. Models should ensure that the external efficacy metric matches the internal falls incidence metric
  7. fLike note 5, this again only concerns decision models using external efficacy evidence. The fall type (e.g., hospitalised fall, fall-induced fracture) for the efficacy data should match that for the model incidence
  8. gDurability of intervention efficacy should not extend beyond the timespan of the intervention study unless the intervention receipt is sustained [22]
  9. hDecision models should ensure that the characteristics of the external intervention study’s target population/sample (e.g., inclusion/exclusion criteria) match those of the model population
  10. iLifetime horizon is recommended by the expert guideline on falls prevention economic evaluation [22]
  11. jAn example of a method used to characterise the dynamic complexity of falls risk is to incorporate tunnel states in Markov cohort models to capture the secular age-related increase in falls risk [28]
  12. kProspective reduction in structural uncertainty can be achieved through stakeholder engagement and model conceptualisation that precedes model parameterisation [15]
Back to article page

BMC Health Services Research

ISSN: 1472-6963

Contact us