There is growing interest in the actions that cities are taking to improve health and healthcare. This is in keeping with the increasing importance of cities in determining global population health. The earliest initiative was the House of Lords Select Committee Report on Metropolitan Hospitals, Provident and Other Public Dispensaries and Charitable Institutions for the Sick Poor,(43) published in 1892. Whilst the social context and healthcare delivery have developed significantly, the issues identified in this report, including the need for co-ordination between services and unacceptable inequalities, remain as relevant today. The collective body of evidence identified in this review, built up over the last 130 years, has hitherto been poorly documented and characterised. This may in part explain why longstanding issues remain pertinent today.
Education was the most frequently applied type of initiative (26%), with the trend increasing over time. It consistently features as one of the most reported initiatives in the last 15 years. With non-communicable diseases – often referred to as ‘life style choices’(44, 45) - being such a high proportion (92%) of the subject areas that initiatives were seeking to address, it could be expected that education features so strongly. Education initiatives tend to be straightforward to write up and evaluate(46), which lends them towards publication. They are also commonly associated with funding(47) that requires some kind of evaluation or report, which again lends towards publication. However, in our experience, trying to change individual behaviour without changing the system or environment within which people live does not work. The term ‘life style choices’ is a misnomer. The reality is that we are strongly influenced by explicit and implicit prompts in our environment and those with the least resources have the least choice and are the most lectured.
Changes to the physical environment is linked with to changing the environment within which we live. It was therefore disappointing that these types of initiatives accounted for only 21% (n = 336) of the total. It was also disappointing that only 21 of the 336 reported an outcome measure. These were addressing issues of obesity (48),(49) smoking(50),(51),(52) water borne infections(53),(54),(55) and traffic related accidents and active travel.(56),(57),(58),(59, 60) These changes to physical environment were instigated by policy or legal changes in just under half (43%) of the initiatives. In considering the balance between education and system or environmental initiatives, more focus and effort should be on system and environment to ‘make the healthy choice the easy choice’.
Service reform (23% of the total citations) performed best on reporting an outcome measure (n = 34). These ranged from centralising heart attack and stroke services(61),(62) injection drug user facilities for harm reduction(63) decentralising and creating networks for care for diabetes(64) and dementia(65) to mass nicotine replacement therapy.(66) Service reform, which is defined as ‘a change or addition to a health service’, is akin to changes to the physical environment or system within which we live. It is also most directly linked to health services. Therefore, service reforms have a direct and clear link to any changes in outcomes. This is not to say the changes are easy or straightforward. Changing any provision of health service, particularly if controversial services or moving services further from a patient’s home, is likely to understandably attract public and political interest. Clinical leadership and genuine (public and political) engagement are key.
A relatively small number (n = 124) of initiatives described what was believed to be a comprehensive suite of initiatives. City health departments were most notable in leading these in a diverse range of cities including New York City (USA)(67), Geelong (Australia)(68) Barcelona (Spain)(69), Taipei (Tiawan)(70) to Nizwa (Oman)(71). Coincidently, there were also 124 records that stated an outcome measure. It is striking that although a comprehensive approach only represented 8% of the total, they represent almost a quarter (23%) of the number with an outcome measure. There could be a number of reasons why outcome measures feature strongly when a comprehensive suite of initiatives is undertaken. It could be seen how a range of initiatives addressing the same subject area would be more likely to bring about a change in outcome compared to a single initiative.
City Health Departments are strongly associated with City Mayors. However, of the 1614 citations only 44 (3%) referenced a Mayor as having any kind of role. Considering Mayors are such features of cities, this was surprising. It was also notable, in general, that initiatives across all areas appeared quite piecemeal and carried out in insolation; a single initiative linked to a single health issue or risk factor without reference to a wider strategy, aim or other supporting initiatives. This could simply be a product of the way the papers were written. Indeed, there is a neat simplicity to this approach. However, the reality of the real world is that a single initiative is unlikely to making meaningful and lasting impact on long-standing issues. Cities are a meta-system with population health dictated by multiple interactions and multidirectional feedback loops. A change to one part of the system could both result in multiple unintended consequences or no impact on outcomes at all. Improving health in cities requires multiple actions, by multiple actors, at multiple levels. No single organisation, sector or initiative can solve the complex interlinked issues. The political and system leadership across a collaboration of partners by a Mayor will be crucial. The role and impact of Mayors in city residents’ health would warrant further investigation.
The initiative itself also affected how it was instigated. It should be noted that “not stated” featured highly across all groups, ranging up to 54% in education and support and totalling 529 (33%) across all initiatives. Therefore, care should be taken with interpreting this result. The comprehensive initiatives were broadly split one fifth policy, collaboration and city health department, with 20%, 18%, 27%, respectively. Education initiatives were most associated with collaboration (15%), city health department (9%) and policy (6%). Changes made to the physical environment were strongly associated with policy (22%) and law (21%). Digital, IT and data was most commonly instigated by city health departments (21%).
Table 3: Type of initiative and how it was instigated
The majority of papers were from North America (57% of total, almost double the total of Europe) and in particular New York City which alone accounts for 17% (n = 279) of the total, more than the entire WHO Healthy City output found. It is not clear if this is because New York City is taking significant steps to improve health and healthcare or they are publishing a lot of their work, or both. This features in the bias and limitations identified in the section below. It also highlights the opportunity for further study through a survey of city health departments to get a more complete picture of the initiatives cities are undertaking to improve population health.
The breadth of subject areas being addressed from a number of different perspectives and the interrelated nature of the subject areas is striking. For example, diet is known to impact on obesity,(72) and cancer(73), and cardiovascular(74) disease and obesity is known to impact on cancer,(75) cardiovascular,(76) mental health(77) and mental health is known to impact on obesity.(78) Therefore, an initiative to improve diet could be an initiative to address obesity, cancer, cardiovascular disease, mental health or health generally.
The complexity of a range of subject areas being addressed could be simplified when the major causes of chronic diseases are well known. Indeed, the four main risk factors (tobacco use; harmful alcohol use; physical inactivity and unhealthy diet) account for at least 80% of all heart disease, stroke and type 2 diabetes and 40% of cancers.(79) The authors would add a fifth risk factor; planetary health. Planetary health is widely acknowledged to be a significant risk to human health(80),(81),(82) particularly impacting cities.(83, 84) The health risk factors are also directly linked to planetary health factors, for example diet(85) and evidence shows polices to promote sustainable food supply can improve personal and planetary health.(86) The authors would call these ‘The Vital 5’. They are the 5 most important health risk factors. These ‘Vital 5’ are most concentrated in deprived areas(87) and show the greatest increase is in low and middle income countries, even faster than has historically occurred in high income countries.(88) Therefore, action on these ‘Vital 5’ will be action on inequalities. Every city should develop a comprehensive strategy and action plan to address these ‘Vital 5’ of tobacco use; harmful alcohol use; physical inactivity, unhealthy diet and planetary health
The 20th century was defined by breakthroughs in communicable diseases. Communicable diseases are certainly not eradicated, HIV(89),(90),(91) and TB(92),(93),(94) can still be seen to be particular issues in cities. The Covid-19 pandemic has also disproportionately impacted cities.(12) However, Covid-19 has also reminded us of the importance of these Vital 5. It is known that smoking,(95) alcohol,(96) poor diet,(97) physical inactivity(98) and air quality(99), all contribute to worse outcomes of Covid-19. They also contribute to other factors which have worse outcomes of Covid-19, such as high blood pressure(100) and obesity.(101) Crucially, it is also know that damaging the environment will make future pandemics more likely.(102),(103) If cities were to focus on these Vital 5 they would be improving immediate and long term population health.
This study has several limitations. The deliberate decision was taken to be as broad and inclusive as possible in the papers included. Mays et al(104) and Strech and Tilburt(105) identify the tension between using only academically rigorous, but therefore less, research and the inclusion of other sources of data, which increase the volume, but that are more susceptible to bias and therefore risk contaminating results. Because of the volume of papers and restrictions on time only the title and abstract were used in the data extraction. Formal quality assessment of the studies was not undertaken. Only studies in English language were included. A large volume of citations were from one city (New York City) which could introduce bias. Finally, given the heterogeneity of the study designs and data sources meta-analysis was not undertaken. They authors acknowledge that despite a comprehensive and systematic approach they may have missed relevant documentation.
The authors believe non-communicable diseases will be the issue of the 21st century, accounting for over 40 million annual deaths globally (72% of total deaths).(106) The Vital 5; tobacco use; harmful alcohol use; physical inactivity, unhealthy diet and planetary health impact on both non-communicable diseases and the likelihood of future communicable pandemics. The progress made in the treatment and prevention of HIV from death sentence 20 years ago to entirely treatable and preventable long-term condition, plus the rapid response to Covid-19 demonstrates what could be achieved to eliminate these ‘Vital 5’ with collective determination and resources. Cities have demonstrated they are well placed to take this action.