Doctor discussing lung scan results with patient
VoxEU Column Health Economics

The role of social effects on lung cancer treatment and innovation adoption

Lung cancer is strongly associated with smoking. Despite recent advances in available therapies, it is still characterised by low treatment rates and a lack of research funds. This column presents evidence on the impact of social effects on treatment and innovation adoption. Social effects deter access to treatment: reducing social discrimination would increase treatment rates by 4% and the use of innovative therapies by 3%, while social effects account for around 2% of the gap in research funding for lung cancer.

Lung cancer is the most commonly diagnosed cancer and the leading cause of cancer-related deaths worldwide. It accounts for 13% of all new cancer cases and has the lowest survival rate among leading cancers. Fortunately, the advent of targeted and immunotherapy agents has revolutionised our understanding of the disease in the past decade (Lichtenberg 2019). These new therapies significantly improve patient survival, are often administered orally (instead of intravenously), and are associated with milder side effects.

Unfortunately, and surprisingly, many patients have not taken full advantage of these innovations: lung cancer patients access treatment at much lower rates than patients affected by cancers with similar (untreated) survival rates. These striking differences in adoption are not fully explained by heterogeneity in the diseases or patients (Sacher 2015).

Social effects in lung cancer and R&D funding

One explanation for the lack of adoption lies in the nature of the disease itself and, more specifically, the negative social effects associated with having lung cancer. Most lung cancer patients (around 80%) have a smoking history, and the strong association of lung cancer with smoking can result in biased beliefs and stigma connected with the disease. Patients incorrectly believe that therapy is ineffective (biased beliefs) or they feel shame about having lung cancer, as conferred by the social representation of lung cancer as being self-inflicted (stigma). In addition, as biased beliefs and stigma constitute barriers to accessing treatment, they may also hinder the adoption and diffusion of innovative therapies for cancer patients.

In turn, a lower number of treated patients impacts the number and value of investments made in innovative therapies (Acemoglu and Linn 2004, Dubois et al. 2015). While lung cancer is responsible for 32% of cancer deaths, it receives only 10% of cancer research funding. Kamath et al. (2019) report an average spending of $2,229 in research per lung cancer death, compared to $24,442 for breast cancer.

In a recent paper (Grigolon and Lasio 2023), we quantify to what extent social effects (biased beliefs and stigma) hinder access to treatment, the adoption of innovative therapies, and investments in R&D. While the current literature has explored a variety of motives to explain heterogeneity in adoption patterns, from learning and uncertainty about side effects to healthcare culture, we are the first to explore the connection between disease stigmatisation and innovation.

We combine administrative micro-level datasets for the population of patients diagnosed with lung cancer in the Canadian province of Ontario between 2008 and 2018. We explain our contribution in two steps:

  • We first identify the role of social effects on the probability of treatment.
  • As social effects deter treatment, we then quantify the impact of the resulting lower market size (number of treated patients) on R&D funding.

The role of social effects on the probability of treatment

The share of untreated patients living in the same neighbourhood is our measure of social effects, as it exploits the granular geographic information available in the data and captures the role of a patient’s reference group in the decision to seek treatment. A survey of 400 adults across Ontario, which we conducted within our study, shows that around 20% of respondents feel ‘low sympathy’ for a person with lung cancer, compared to sympathy felt towards people with other tumours. The variation in the degree of stigma across communities in Ontario positively correlates with the measure we construct in our data.

Causal social effects are difficult to identify empirically because of simultaneity and correlated effects (Manski 1993). We address simultaneity by focusing on the choice of newly diagnosed patients, whose decision to pursue treatment may be influenced by patients from the same neighbourhood diagnosed in previous years, but not vice versa.

To disentangle social effects from correlation in unobserved attributes, we isolate the variation in treatment choices of fellow patients living in the same community, independent of the unobservables. In particular, we rely on the quasi-random variation in treatment rates of the reference group: in the Ontario universal healthcare system, patients do not access secondary care directly and do not have the option to choose their oncologist. Furthermore, clinicians work in regional cancer centres and do not have ties to a specific neighbourhood.

We construct the (risk-adjusted) average treatment propensity of physicians treating patients in the reference group in previous years and use it as an instrumental variable for treatment rates in the neighbourhood. Placebo tests using other cancer types (for which social effects are less of a concern) confirm the effectiveness of our identification strategy.

We find that placing all patients in a neighbourhood characterised by low social stigma decreases the share of untreated patients by 4%. It also increases the use of innovative therapies by 3%. Treating those additional patients implies additional costs, but these costs are vastly outweighed by the benefits in terms of gains in survival (Sun et al. 2010, Conti et al. 2015, Van Nieuwerburgh and Koijen 2018).

R&D investment and market size

In the second step, we relate R&D investments for different cancers to market size, measured by the number of treated patients. Our estimated elasticity suggests that a 10% increase in market size is associated with a 3.4 to 5.6% increase in R&D spending. Back-of-the-envelope calculations indicate that by discouraging treatment, social stigma and biased beliefs are responsible for around 2% of the gap in research funding for lung cancer with respect to other common cancers. This amounts to $7 million every year in US public funding alone.

Concluding remarks: Do social effects prevent the adoption of innovative treatments?

Our empirical results inform the policy debate on lung cancer stigma and improve the wider societal understanding of lung cancer. We also offer strong evidence showing that patients face accessibility problems linked to stigma, which then slow the adoption of innovative treatments and lower the incentives to invest in R&D.

We explore and quantify the link between social discrimination, adoption of innovation, and R&D investment. Recent works have investigated the role of social stigma in learning and reporting the status of stigmatised diseases, such as HIV or mental health issues (Thornton 2008, Yu 2019, Bharadwaj et al. 2017, Cronin et al. 2020).

Future research on stigmatised diseases, for which scientific knowledge has produced significant therapeutic advances, will be helpful in understanding to what extent societal biases hinder the diffusion of innovation and, in turn, discourage further R&D investments.


Acemoglu, D, and J Linn (2004), “Market size in innovation: Theory and evidence from the pharmaceutical industry”, Quarterly Journal of Economics 119(3): 1049–90.

Bharadwaj, P, M M Pai, and A Suziedelyte (2017), “Mental health stigma”, Economics Letters 159: 57–60.

Conti, R, D Howard, and E Berndt (2015), “Cancer drug prices rise with no end in sight”,, 25 March.

Cronin, C J, M P Forsstrom, and N W Papageorge (2020), “What good are treatment effects without treatment? Mental health and the reluctance to use talk therapy”, NBER Working Paper 27711.

Dubois, P, O de Mouzon, F Scott-Morton, and P Seabright (2015), “Market size and pharmaceutical innovation”, RAND Journal of Economics 46(4): 844–71.

Grigolon, L, and L Lasio (2023), “Biased beliefs and stigma as barriers to treatment and innovation adoption”, CEPR Discussion Paper 17938.

Kamath, S D, S M Kircher, and A B Benson (2019), “A comparison of cancer burden and funding from the National Cancer Institute and nonprofit organizations reveals disparities in the distribution of funding across cancer types”, Journal of the National Comprehensive Cancer Network 17(7): 849–54.

Lichtenberg, F (2019), “The impact of biomedical innovation on cancer mortality”,, 8 February.

Manski, C F (1993), “Identification of endogenous social effects: The reflection problem”, Review of Economic Studies 60(3): 531–42.

Sun, E, C Reyes, T Philipson, D Lakdawalla, A Jena, and D Goldman (2010), “An economic evaluation of the war on cancer”,, 11 January.

Thornton, R L (2008), “The demand for, and impact of, learning HIV status”, American Economic Review 98(5): 1829–63.

Van Nieuwerburgh, S, and R Koijen (2018), “Financing the war on cancer”,, 1 July.

Yu, H (2019), “Social stigma as a barrier to HIV testing: Evidence from a randomized controlled trial in Mozambique”, working paper.

377 Reads