Abstract
The health benefits of spas have been hypothesized for centuries. If this hypothesis is correct, spa therapy offers a low cost alternative to more expensive and potentially more invasive medical treatments for ailments such as back pain and arthritis. We use individual-level panel data to isolate the effect of spa therapy on missed workdays and hospital visits in Germany. Simple correlations suggest a self-selection bias — spa visits are associated with increased absenteeism and hospitalization. However, when we exploit the longitudinal nature of the data, we find that spa therapy leads to a statistically significant reduction in both absenteeism and hospitalization, though it is not clear if these health benefits justify the cost of spa therapy.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
1. Although there are technical distinctions recognized by some health researchers, we treat spa therapy, balneotherapy, and hydrotherapy as effectively interchangeable.
2. Inpatient treatments provide a much more controlled environment for the patient than outpatient treatments. For example, no alcohol consumption is allowed when an individual receives an inpatient treatment, and the treatment facility also has control over the diet. This is not the case for an outpatient treatment, where the individual receives treatment for only a few hours a day. The inpatient treatment is a much more comprehensive treatment option. During inpatient treatments patients typically have a program to follow from the morning until the evening, with different treatments every hour, including exercise. Unfortunately, we cannot test directly whether the health benefit of an inpatient treatment equals that of a non-spa vacation because we do not have data to construct a specific control group for people taking a non-spa vacation. However, people who took a non-spa vacation are included in our data and are thus part of our control group. The reason that we believe that many vacationers are included in our control group is because in Germany workers often receive over 4 weeks of annual vacation.
3. Although we do not present the diagnostic statistics, in all negative binomial models, we reject the hypothesis of no overdispersion, which is why we present negative binomial results instead of the corresponding Poisson model results.
4. Strictly speaking, the fixed effect model for the negative binomial regressions allows for individual dispersion coefficients rather than a vector of individual dummy variables, but the intuition is similar to that for the OLS fixed effect model.
5. Income is potentially endogenous, if, for example, workday loss leads to lower productivity and thus income.
6. We added gender and marital status to our OLS regressions and the point estimate on spas was virtually unchanged, although both male and marital status indicators were statistically significant.
7. We view the inclusion of this firm size variable as a robustness check with respect to the specification of the model. Health insurance and spa treatments are regulated at the federal government level, and are not included in compensation packages of most German firms. Thus there is less reason to include firm size effects as there would be in the U.S. context.
8. The average number of observations for each individual is 4; the minimum is 2 and the maximum number of observations is 10.
9. It is appropriate to interpret the coefficients of the negative binomial model as semi-elasticities when the coefficient are small (β<0.1 or β−0.1), but they need to be transformed for larger values such as that in Table 3. Using exp(β)−1 to calculate the semi-elasticity, the elasticities for Table 3, column 3 are lagged spa −0.0117, age −0.105, and for the lagged spa variable in Table 3, column 4 −0.437.
References
Allard, P., J. Deligne, V. Van Bockstael, and B. Duquesnoy . 1998. Is Spa Therapy Cost-Effective in Rheumatic Disorders. Revue du Rhumatisme (English Edition). 5 (3): 173–180.
Coccheri, S., G. Nappi, M. Valenti, F. Di Orio, E. Altobelli, and S. DeLuca . 2002. Changes in the Use of Health Resources by Patients with Chronic Phlebopathies after Thermal Hydrotherapy. Report from the NAIADE Project, a Nation-Wide Survey on Thermal Therapies in Italy. International Angiology, 21 (2): 196–200.
Constant, F., J.F. Collin, F. Guillemin, and M. Boulange . 1995. Effectiveness of Spa Therapy in Chronic Low Back Pain: Randomized Clinical Trial. Journal of Rheumatology, 22 (7): 1315–1320.
Ekmekcioglu, C., G. Strauss-Blasche, J. Feyertag, N. Klammer, and W. Marktl . 2000. The Effect of Balneotherapy on Ambulatory Blood Pressure. Alternative Therapies in Health and Medicine, 6 (6): 46–53.
Ernst, E., and M.H. Pittler . 1998. Wie effektiv is die Kur? Eine systematische Übersicht andomisierter Studien. Deutsche Medizinische Wochenzeitschrift, 123: 273–277.
Fioravanti, A., M. Valenti, E. Altobelli, F. Di Orio, G. Nappi, A. Crisanti, L. Cantarini, and R. Marcolongo . 2003. Clinical Efficacy and Cost-Effectiveness Evidence of Spa Therapy in Osteoarthritis. The Results of the ‘Naiade’ Italian Project. Panminerva Medica, 45 (3): 211–217.
Konrad, K., T. Tatrai, A. Hunka, E. Vereckei, and I. Korondi . 1992. Controlled Trial of Balneotherapy in Treatment of Low Back Pain. Annals of the Rheumatic Diseases, 51 (6): 820–822.
Leibetseder, V., G. Strauss-Blasche, F. Holzer, W. Marktl, and C. Ekmekcioglu . 2004. Improving Homocysteine Levels Through Balneotherapy: Effects of Sulphur Baths. Clinica Chimica Acta; International Journal of Clinical Chemistry, 343 (1–2): 105–111.
Nguyen, M., M. Revel, and M. Dougados . 1997. Prolonged Effects of 3 Week Therapy in a Spa Resort on Lumbar Spine, Knee, and Hip Osteoarthritis: Follow-Up After 6 Months. A Randomized Controlled Trial. British Journal of Rheumatology, 36 (1): 77–81.
Ohtsuka, Y., N. Yabunaka, I. Watanabe, H. Noro, and Y. Agishi . 1996. Balneotherapy and Platelet Glutathione Metabolism in Type II Diabetic Patients. International Journal of Biometeorology, 39 (3): 156–159.
Pohlmeier, W., and V. Ulrich . 1995. An Econometric Model of the Two-Part Decisionmaking Process in the Demand for Health Care. Journal of Human Resources, 30 (2): 339–361.
Porter, Roy . 1990. The Medical History of Waters and Spas. London: Wellcome Institute for the History of Medicine.
Riphahn, Regina T., Achim Wambach, and Andreas Million . 2003. Incentive Effects in the Demand for Health Care: A Bivariate Panel Count Data Estimation. Journal of Applied Econometrics, 18 (4): 87–405.
Strauss-Blasche, G., C. Ekmekcioglu, N. Klammer, and W. Marktl . 2000. The Change of Well-Being Associated with Spa Therapy. Research in Complementary and Natural Classical Medicine, 7 (6): 269–274.
Szucs, L., L. Ratko, T. Lesko, I. Szoor, Gy Genti, and G. Balint . 1989. Double Blind Trial on the Effectiveness of the Puspokladany Thermal Water on Arthrosis of the Knee Joint. Royal Society of Health Journal, 109: 7–9.
Van Tubergen, A., and S. van der Linden . 2002. A Brief History of Spa Therapy. Annals of Rheumatic Diseases, 61: 273–275.
Verhagen, A.P., H.C. de Vet, R.A. de Bie, A.G. Kessels, M. Boers, and P.G. Knipschild . 1997. Taking Baths: The Efficacy of Balneotherapy in Patients with Arthritis. A Systematic Review. Journal of Rheumatology, 24 (10): 1964–1971.
Wigler, I., O. Elkayam, D. Paran, and M. Yaron . 1995. Spa Therapy for Gonarthrosis: A Prospective Study. Rheumatology International, 15 (2): 65–68.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Klick, J., Stratmann, T. Do Spa Visits Improve Health: Evidence From German Micro Data. Eastern Econ J 34, 364–374 (2008). https://doi.org/10.1057/palgrave.eej.9050038
Published:
Issue Date:
DOI: https://doi.org/10.1057/palgrave.eej.9050038