Do health partnerships with organisations in lower income countries benefit the UK partner? A review of the literature

Our systematic review comprised both peer-reviewed and grey literature, and was conducted in November 2012. Grey literature was included because there are few peer-reviewed papers in this area, and unpublished policy documents and project reviews are likely to play a key part in this emerging and complex field [7‐9].

The inclusion criteria were agreed by the project team on the basis of a preliminary review of definitions of health partnerships. This suggested that health partnerships might impact at individual, institutional and national levels and impacts at all levels were included to gain as thorough an evaluation as possible.

Our inclusion criteria were published or grey literature on health partnerships where:

For each element of the search one team member reviewed all initial results, to ensure that they met the inclusion criteria and 20% of each set of results were cross-checked by a different team member. Any disagreements about application of criteria or extraction of data were resolved through negotiated consensus.

All documents that met the inclusion criteria were scanned in their entirety, and all text that referenced benefits or costs at any of our three levels (individual, institutional and national) was extracted for analysis.

Since the majority of studies found during the search were qualitative in nature, thematic analysis was chosen to synthesise the findings. Thematic synthesis is a method frequently used to analyse data in primary qualitative research, which has three stages which may overlap: the coding of text ‘line-by-line’; the creation of ‘descriptive themes’; and the generation of ‘analytical themes’ [11].

All textual data from websites and papers was therefore extracted into an excel spreadsheet and separated into ‘line-by-line’ content. Through a process of reading and re-reading we generated a comprehensive list of codes to map all recorded costs and benefits to fully describe the data in these terms. The individual costs and benefits were then grouped into descriptive themes that grouped similar outcomes. To understand the implications of this description of costs and benefits we interrogated our data further to draw out their implications for professional development and potential impact on clinical practice and service development. In this process we attempted to interpret and ‘go beyond’ initial data to form new constructs, explanations or hypothesis [12]. The decisions in all, but particularly the third stage, are dependent on the judgement and views of the reviewer, so decisions made at all three stages were discussed and agreed by all team members.

Since we considered benefits at three levels (individual, institutional and national) we sought frameworks to show how these were linked. To link individual benefits to institutional benefits we mapped those identified from the literature review against five frameworks for skill acquisition in the NHS. These frameworks were chosen to cover professional development of doctors of all specialties (the Academy of Royal Colleges’ [13] and NHS Leadership and Development frameworks [14]) and all other NHS workers (the Knowledge and Skills framework [15]), and to cover all stages of careers, from minimum standards at qualification (the General Medical Council’s Good Medical Practice [16]), to skills expected at consultant level and beyond (The Continuing Professional Development curricula [17]). The focus upon frameworks relating to medical careers was chosen as all partnerships studied had included medical professionals, whereas no other health professional was universally represented.

To link individual benefits for health professionals to outcomes for those using UK health services we used a modified version of a framework developed by Wales for Africa [18].

We completed a thematic analysis of the qualitative data from our review, the results of which can be seen in Additional file 1: Appendix 5. The majority of our sources (95%) cited benefits and we developed 40 independent initial codes of benefits to individuals.

These benefits to individuals were assigned to seven domains, six of which were skills-based: clinical skills, management skills, communication and teamwork, patient experience and dignity, policy, academic skills, and the last of which related to personal satisfaction and interest, as shown below in Table 3. Additional file 1: Appendix 6 categorises these individual benefits as knowledge, attitudes and skills.

The evidence for benefit on institutional level, and national level in particular was weak. However, we extracted 10 benefits to institutions, and 10 national benefits. Some of these (such as reputational development and demonstrating Corporate Social Responsibility) seemed to arise directly from the existence of a link, independently of the benefits derived by individuals participating in the link, whilst others (such as staff with an understanding of the global context) seemed to arise as a result of individuals who were part of the institution or national workforce participating in links. The number of individuals participating in a link is very small percentage of the workforce. Therefore the perception that benefits such as “reduction of waste within the NHS” arise is based on the (untested) hypothesis that link participants are able to significantly change institutional and national systems on their return, in order to impact upon service outcomes. 13 sources (32%) cited costs and we developed independent initial codes for costs which were grouped into 5 domains; financial, loss of staff, reputational, health and security and opportunity costs, as demonstrated in Table 4. Many of these costs, particularly with regards to loss of staff and the opportunity costs, arose from the adhoc relationship of partnerships work to other professional activities; for example the difficulties of organising cover for those participating in links was prominently cited as a cost of volunteering.

Some individual benefits such as improved motivation, team-working, and an understanding of patients from different cultures and backgrounds were considered to contribute to UK health workforce development. To investigate this link we mapped the benefits found in the six skills-based domains against five frameworks used by institutions to structure workforce skills and development (building on previous work in this field [19]). The results of this process are shown in Table 5 which documents the close relationships between the benefits which our review suggests arise from participating in health links and the attributes that UK health institutions are seeking to develop within the health care workforce.

The Knowledge and Skills framework, of which core dimensions are considered in Table 5 is further expanded in table Additional file 1: Appendix 7 to detail 29 skill dimensions each with four levels of descriptor. Benefits from health partnerships engagement were shown to map directly onto the most advanced descriptor of each dimension. This suggests any member of the NHS workforce could gain or improve skills from volunteering through a health link in a way which is recognised to be of benefit to individuals in their NHS jobs, institutions and ultimately patient care.

To illustrate this relationship further and to extrapolate it to health care delivery we further developed a framework used by Wales for Africa and published on their website [18]. This shows how the experience of volunteering could impact on individuals, institutions and health care service users, in all six skills-based domains as shown in Figure 1. However, not all new skills were relevant to UK healthcare delivery; such as learning to cope without the imaging equipment available in the UK.

Key findings included:

● There is little peer-reviewed and grey literature published on this topic, and that which has been published largely fails to meet high standards of formal academic rigor.

● Much of the literature is written by those involved in facilitating or advocating for links, with few external evaluations.

● With the limitations described above, the existing literature suggests that the benefits of health partnerships outweigh costs.

● The individual benefits fell into 7 domains, 6 of which were skills-based, and are closely linked to existing UK frameworks to structure health workforce development.

● We extracted 10 benefits to institutions and 10 national benefits.

● Institutional and national benefits could be seen to arise both from the existence of links, and as a result of individuals from the UK workforce benefitting through link experiences.

● There is a theoretically clear trajectory that links volunteer experience to improved health service delivery for some categories of benefits (e.g. leadership) but not others (e.g. working in situations without access to clinical investigations).

● Five cost domains were identified: financial, loss of staff, reputational, health and security and opportunity costs.

● More data is needed on the relative impact of costs and benefits for individual volunteers or within an individual link, and the factors influencing these.

We reviewed the limited literature available on the UK benefits of health partnerships and identified a range of costs and benefits. There was wide variation between the links reviewed, making it difficult to draw universal comparable conclusions. Standardised, valid and reliable evaluation tools should be used in the future to enable comparison between links, and to facilitate aggregation of data between different trips within a partnership, and different partnerships.

Our analysis showed that the impact of a link is closely related to how effectively it is run, and this was a key confounding factor. For example, the cost of a negative perception of the UK institutions or UK was raised by southern partners [20, 21] who commented that where links are poorly run or have prominent power imbalances they have in some instances harmed the UK reputation. However, where run appropriately links have been seen to enhance the reputation of the UK; this variety within links and their impacts was noted in the Department of Health’s evaluation of links (2008) [5].

We have shown that the skills developed through volunteering in a link are considered valuable skills for NHS employees according to numerous professional development frameworks. However, it is impossible to map the level of skills attained against the level required in these frameworks as existing studies rely heavily on volunteers reports of subjective improvement, untested against such frameworks. Thus, further work is required to develop a structure for optimising goal setting and measuring skill development during health partnership placements.

The literature suggests that health partnerships could have an important positive impact for UK health professionals and institutions. However some skills learnt during volunteering are more transferable than others, or transferable in some contexts but not others. For example, experience of tropical diseases might be only relevant to those who work in the UK with populations where such diseases are common. More careful analysis of the factors affecting translation of experience to impact on clinical practices and services at each stage is required in order to modify them so as to maximise the positive impacts of health partnerships to the UK. Some possibilities have been shown in Figure 1.

The findings of this review and their application to key frameworks suggested strategies with potential to maximise the benefits of health links. Since many costs appear to result from the ad hoc relationship between partnership work and other professional activities one such strategy would be the formal integration of volunteering within partnerships into NHS Continuing Professional Development (CPD) frameworks (as previously explored by Longstaff B). This would potentially release study leave funding to support volunteering within links, minimising additional costs to individuals and the NHS, making it easier to organise cover, and therefore reducing many of the ‘loss of staff’ and ‘financial’ costs raised. Volunteering at least partially within study leave would protect volunteers’ annual leave, and thus help prevent burnout. Benefits of the strategy would include transferring some financial burden from the individual to the institution, which might in turn increase engagement, formal recognition of the skills gained by individuals through link participation, and justification of investment of NHS resources in partnerships. The Welsh NHS has already taken this step [22] and the impact of this scheme needs thorough evaluation including cost effectiveness analysis.

More data is needed to quantify the benefits identified and to balance these against the costs. Quantification of benefits to date [5] is entirely based on self-reporting, and on ordinal scales, providing weak grounds for comparison. Costs have only been evaluated in the financial domain and estimates of the cost of the trips overseas for an average link, vary from £10,000 - £20,000 per annum [1] to significantly higher figures [5]. Even where objective measures of costs (or disadvantages) do exist, it is challenging to compare these between the different domains identified in our analysis.

Our research clearly demonstrated the heterogeneous nature of health partnerships. New standardised evaluation tools should distinguish between the different benefits and costs experienced by different health professionals with different levels of experience in different links, thus allowing appropriately targeted future funding and personal support.

Comparison of engagement in links to alternative methods of health workforce development would enable effective comparison of learning outcomes and costs to the institution. Thus future attempts at evaluation should aim to gather baseline measures prior to engagement in a link, and to incorporate more objective, independently-assessed measures of improvement of practice in the UK. These could be based around existing competency frameworks, and existing initiatives [23, 24] should be encouraged and shared. Such attempts should ideally be conducted alongside measuring benefits to the partner country, to reduce the risk of participants viewing volunteering as principally an activity for self-development, potentially at the cost of the ‘recipient’ partner.

There is a real need to develop standardised methods of data collection and reporting in this area (as previously explored by Syed et al.) to enable greater understanding of health innovation diffusion and how developed countries can benefit from the work of developed countries, both in the field of health links and beyond [25].