Drug Safety

, Volume 27, Issue 13, pp 1019–1042 | Cite as

Treatment Costs to Prevent or Treat Upper Gastrointestinal Adverse Events Associated with NSAIDs

A Review
  • Elham Rahme
  • Alan N. Barkun
  • Viviane Adam
  • Marc Bardou
Review Article


The widespread use of nonselective NSAIDs and cyclo-oxygenase (COX)-2 inhibitors has a substantial impact on healthcare budgets worldwide. The cost of their gastrointestinal (GI) adverse effects is a major component of their direct cost and has received much attention in the literature. Published studies have often differed in their methodologies and results. It is important for decision makers to understand the reasons for these differences in order to make informed decisions.

We conducted a literature review to summarise data that evaluate the direct costs of NSAID-related GI adverse effects worldwide. This resulted in 789 articles from which 29 studies met the inclusion criteria and were fully reviewed. Of these 29, the 9 studies that assessed the cost of COX-2 inhibitors were all based on decision economic models, compared with only 7 of the remaining 20 studies, which assessed the cost of nonselective NSAIDs. In most studies, the perspective was that of the healthcare payer and the costs assessed were reimbursement costs. Costs of GI events almost doubled between regular users and non-users of nonselective NSAIDs and were much higher in high-dose versus low-dose users. The ratio of the total cost of nonselective NSAIDs to their acquisition cost reported in all studies varied from 1.36 to 2.12. Both of these numbers were reported in one single study assessing several different NSAIDs in France. Thus, the GI adverse events attributable to nonselective NSAIDs are substantial, and their costs often exceed the cost of the nonselective NSAID itself.

The acquisition cost of the COX-2 inhibitors was the main driver of their total cost. The GI adverse effects with the COX-2 inhibitors added 10–20% to their acquisition cost in North America, while this increase was about 50% in some European countries. Decision analysis models showed that the direct costs of COX-2 inhibitors were lower than those of nonselective NSAIDs in patients at risk of NSAID gastropathy but higher in patients at no to low risk of gastropathy. Thus, from an economic perspective, the healthcare system would benefit from treating patients at risk of NSAID gastropathy with COX-2 inhibitors, but not those at no to low risk.


  1. 1.
    Roth SH. NSAIDs gastropathy: a new understanding. Arch Intern Med 1996; 156(15): 1623–8PubMedCrossRefGoogle Scholar
  2. 2.
    Goldstein JL, Silverstein FE, Agrawal NM, et al. Reduced risk of upper gastrointestinal ulcer complications with celecoxib, a novel COX-2 inhibitor. Am J Gastroenterol 2000; 95(7): 1681–90PubMedCrossRefGoogle Scholar
  3. 3.
    Simon AM, Manigrasso MB, O’Connor JP. Cyclo-oxygenase 2 function is essential for bone fracture healing. J Bone Miner Res 2002; 17(6): 963–76PubMedCrossRefGoogle Scholar
  4. 4.
    Pai R, Szabo IL, Giap AQ, et al. Nonsteroidal anti-inflammatory drugs inhibit re-epithelialization of wounded gastric monolayers by interfering with actin, Src, FAK, and tensin signaling. Life Sci 2001; 69(25-26): 3055–71PubMedCrossRefGoogle Scholar
  5. 5.
    Colville-Nash PR, Gilroy DW. Potential adverse effects of cyclooxygenase-2 inhibition: evidence from animal models of inflammation. Biodrugs 2001; 15(1): 1–9PubMedCrossRefGoogle Scholar
  6. 6.
    Hayllar J, Bjarnason I. NSAIDs, Cox-2 Inhibitors, and the Gut. Lancet 1995; 346(8990): 1629PubMedCrossRefGoogle Scholar
  7. 7.
    Fries JF, Miller SR, Spitz PW, et al. Toward an epidemiology of gastropathy associated with nonsteroidal antiinflammatory drug-use. Gastroenterol 1989; 96(2): 647–55Google Scholar
  8. 8.
    Wallace JL. Nonsteroidal anti-inflammatory drugs and gastroenteropathy: the second hundred years. Gastroenterology 1997; 112(3): 1000–16PubMedCrossRefGoogle Scholar
  9. 9.
    Blower AL, Brooks P, Fenn GC, et al. Emergency admissions for upper gastrointestinal disease and their relation to NSAID use. Aliment Pharmacol Ther 1997; 11: 283–91PubMedCrossRefGoogle Scholar
  10. 10.
    Brooks A, Fenn GC, Hill A, et al. Emergency admissions for upper gastrointestinal disease and their relation to NSAID use. Aliment Pharmacol Ther 1997; 11: 283–91PubMedCrossRefGoogle Scholar
  11. 11.
    Griffin MR. ai]Epidemiology of nonsteroidal anti-inflammatory drug-associated gastrointestinal injury. Am J Med 1998; 104: 23S–9SPubMedCrossRefGoogle Scholar
  12. 12.
    Laine L, Bombardier C, Hawkey CJ, et al. Stratifying the risk of NSAID-related upper gastrointestinal clinical events: results of a double-blind outcomes study in patients with rheumatoid arthritis. Gastroenterology 2002; 123(4): 1006–12PubMedCrossRefGoogle Scholar
  13. 13.
    Silverstein FE, Graham DY, Senior JR, et al. Misoprostol reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs: a randomized double-blind placebo-controlled trial. Ann Intern Med 1995; 123(4): 241–9PubMedGoogle Scholar
  14. 14.
    Walt RP. Misoprostol for the treatment of peptic ulcer and antiinflammatory-drug-induced gastroduodenal ulceration. N Engl J Med 1992; 327(22): 1575–80PubMedCrossRefGoogle Scholar
  15. 15.
    Hawkey CJ, Karrasch JA, Szczepanski L, et al. Omeprazole compared with misoprostol for ulcers associated with nonsteroidal antiinflammatory drugs: Omeprazole verus Misoprostol for NSAID-induced Ulcer Management (OMNIUM) study group. N Engl J Med 1998; 338(11): 727–34PubMedCrossRefGoogle Scholar
  16. 16.
    Dammann HG, Walter TA, Hentschel E, et al. Famotidine: proven once-a-day treatment for gastric ulcer. Scand J Gastroenterol 1987; 134: 29–33CrossRefGoogle Scholar
  17. 17.
    Hawkey CJ. Progress in prophylaxis against nonsteroidal antiinflammatory drug-associated ulcers and erosions. Am J Med 1998; 104: 67S–74SPubMedCrossRefGoogle Scholar
  18. 18.
    Taha AS, Hudson N, Hawkey CJ, et al. Famotidine for the prevention of gastric and duodenal ulcers caused by nonsteroidal antiinflammatory drugs. N Engl J Med 1996; 334(22): 1435–9PubMedCrossRefGoogle Scholar
  19. 19.
    Candelli M, Carloni E, Armuzzi A, et al. Role of alfate in gastrointestinal diseases. Panminerva Med 2000; 42: 55–9PubMedGoogle Scholar
  20. 20.
    Simon LS. Role and regulation of cyclooxygenase-2 during inflammation, position paper by expert panel on COX-1/COX-2. Am J Med 1999; 106(5B): 37S–42SPubMedCrossRefGoogle Scholar
  21. 21.
    Ormrod D, Wellington K, Wagstaff AJ. Valdecoxib. Drugs 2002; 62(14): 2059–71PubMedCrossRefGoogle Scholar
  22. 22.
    Harris SI, Kuss M, Hubbard RC, et al. Upper gastrointestinal safety evaluation of parecoxib sodium, a new parenteral cyclooxygenase-2-specific inhibitor, compared with ketorolac, naproxen, and placebo. Clin Ther 2001; 23(9): 1422–8PubMedCrossRefGoogle Scholar
  23. 23.
    Cochrane DJ, Jarvis B, Keating GM. Etoricoxib. Drugs 2002; 62: 2637–51PubMedCrossRefGoogle Scholar
  24. 24.
    Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS study: a randomized controlled trial. Celecoxib long-term arthritis safety study. JAMA 2000; 284(10): 1247–55PubMedCrossRefGoogle Scholar
  25. 25.
    Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis: VIGOR study group. N Engl J Med 2000; 343(21): 1520–8PubMedCrossRefGoogle Scholar
  26. 26.
    Chan FKL, Hung LCT, Suen BY, et al. Celecoxib versus diclofenac and omeprazole in reducing the risk of recurrent ulcer bleeding in patients with arthritis. N Engl J Med 2002; 347(26): 2104–10PubMedCrossRefGoogle Scholar
  27. 27.
    Rahme E, Marentette MA, Toubouti Y, et al. Are rofecoxib and celecoxib associated with fewer gastrointestinal complications compared to conventional NSAIDs in real-life clinical practice? [abstract]. Arthritis Rheum 2002; 466(9): S118Google Scholar
  28. 28.
    Langman MJ, Jensen DM, Watson DJ, et al. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA 1999; 282 (20): 1929–33CrossRefGoogle Scholar
  29. 29.
    Brooks P, Emery P, Evans JF, et al. Interpreting the clinical significance of the differential inhibition of cyclooxygenase-1 and cyclooxygenase-2. Rheumatology 1999; 38(8): 779–88PubMedCrossRefGoogle Scholar
  30. 30.
    Hawkey CJ, Laine L, Simon LS. Comparison of the effects of rofecoxib (a yclooxygenase 2 inhibitor), ibuprofen and placebo on gastroduodenal mucosa of patients with osteoarthritis. Arthritis Rheum 2000; 43: 370–7PubMedCrossRefGoogle Scholar
  31. 31.
    Juni P, Rutjes AW, Dieppe PA. Are selective COX-2 inhibitors superior to traditional non-steroidal anti-inflammatory drugs? BMJ 2002; 324(7353): 1538CrossRefGoogle Scholar
  32. 32.
    Editorial staff. Cox-2 agents’ impact on NSAID and GI drug use and cost: what’s been seen in practice? Formulary 2000; 35: 71–5Google Scholar
  33. 33.
    Ray WA, Stein CM, Daugherty JR, et al. COX-2 selective nonsteroidal anti-inflammatory drugs and risk of serious coronary heart disease. Lancet 2002; 360(9339): 1071–3PubMedCrossRefGoogle Scholar
  34. 34.
    Schnitzer T. Cyclogenase-2-specific inhibitors: are they safe? Am J Med 2001; 110(1A): 46S–9SPubMedCrossRefGoogle Scholar
  35. 35.
    Kato M, Nishida S, Kitasato H, et al. Cyclooxygenase-1 and cyclooxygenase-2 selectivity of non-steroidal anti-inflammatory drugs: investigation using human peripheral monocytes. J Pharm Pharmacol 2001; 53(12): 1679–85PubMedCrossRefGoogle Scholar
  36. 36.
    Clinard F, Sgro C, Bardou M, et al. Non-steroidal anti-inflammatory drug prescribing patterns in general practice: comparison of a general practitioner-based survey and a pharmacybased survey in France. Pharmacoepidemiol Drug Saf 2001; 10(4): 329–38PubMedCrossRefGoogle Scholar
  37. 37.
    Rahme E, Marentette MA, Toubouti Y, et al. Profiles of patients who used coxibs, meloxicam, and other nonselective NSAIDs: analysis of the RAMQ database in Quebec, Canada [abstract]. Ann Rheum Dis 2002; 61(S1): 347Google Scholar
  38. 38.
    Clinard F, Bardou M, Sgro C, et al. Non-steroidal anti-inflammatory and cytoprotective drug co-prescription in general practice. Eur J Clin Pharmacol 2001; 57: 737–43PubMedCrossRefGoogle Scholar
  39. 39.
    De Pouvourville G. The economic consequences of NSAIDinduced gastropathy: the French context. Scand J Rheumatol Suppl 1992; 96: 49–53PubMedCrossRefGoogle Scholar
  40. 40.
    Fendrick AM, Bandekar RR, Chernew ME, et al. Role of initial NSAID choice and patient risk factors in the prevention of NSAID gastropathy: a decision analysis. Arthritis Rheum 2002; 47(1): 36–43PubMedCrossRefGoogle Scholar
  41. 41.
    Crott R, Makris N, Barkun A, et al. The cost of an upper gastroduodenal endoscopy: an activity-based approach. Can J Gastroenterol 2002; 16(7): 477–86Google Scholar
  42. 42.
    Bank of Canada. Currency converter [online]. Available from URL: http://www.bank-banque-canada.ca/en/exchform.htm [Accessed 2004 Sep 9]Google Scholar
  43. 43.
    Hochberg MC, Altman RD, Brandt KD, et al. Guidelines for the medical management of osteoarthritis: part II. Osteoarthritis of the knee: American College of Rheumatology. Arthritis Rheum 1995; 38(11): 1535–46PubMedCrossRefGoogle Scholar
  44. 44.
    Hochberg MC, Altman RD, Brandt KD, et al. Guidelines for the medical management of osteoarthritis: part I. Osteoarthritis of the hip. Arthritis Rheum 1995;38(11): 1535–40PubMedCrossRefGoogle Scholar
  45. 45.
    Schnitzer TJ. Update of ACR guidelines for osteoarthritis: role of the coxibs. J Pain Symptom Manage 2002; 23(4S): S24–30PubMedCrossRefGoogle Scholar
  46. 46.
    US Department of Labor, Bureau of Labor Statistics. 2003 consumer price index [online]. Available from URL: ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt [Accessed 2004 Sep 9].Google Scholar
  47. 47.
    Johnson RE, Hornbrook MC, Hooker RS, et al. Analysis of the costs of NSAIDs-associated gastropathy: experience in a US health maintenance organisation. Pharmacoeconomics 1997; 12(1): 76–88PubMedCrossRefGoogle Scholar
  48. 48.
    Simon LS, Zhao SZ, Arguelles LM, et al. Economic and gastrointestinal safety comparisons of etodolac, nabumetone, and oxaprozin from insurance claims data from patients with arthritis. Clin Ther 1998; 20(6): 1218–35PubMedCrossRefGoogle Scholar
  49. 49.
    Smalley WE, Griffin MR, Fought RL, et al. Excess costs from gastrointestinal disease associated with nonsteroidal antiinflammatory drugs. J Gen Intern Med 1996; 11: 461–9PubMedCrossRefGoogle Scholar
  50. 50.
    Kephart G, Sketris I, Smith M, et al. Coprescribing of nonsteroidal anti-inflammatory drugs and cytoprotective and antiulcer drugs in Nova Scotia’s senior population. Clin Ther 1995; 17(6): 1159–73PubMedCrossRefGoogle Scholar
  51. 51.
    Rahme E, Joseph L, Kong SX, et al. Gastrointestinal health care resource use and costs associated with nonsteroidal antiinflammatory drugs versus acetaminophen: retrospective cohort study of an elderly population. Arthritis Rheum 2000; 43(4): 917–24PubMedCrossRefGoogle Scholar
  52. 52.
    Rahme E, Joseph L, Kong SX, et al. Cost of prescribed NSAIDrelated gastrointestinal adverse events in elderly patients. Br J Clin Pharmacol 2001; 52(2): 185–92PubMedCrossRefGoogle Scholar
  53. 53.
    Rahme E, Joseph L, Kong SX, et al. Gastrointestinal-related healthcare resource usage associated with a fixed combination of diclofenac and misoprostol versus other NSAIDs. Pharmacoeconomics 2001; 19(5): 577–88PubMedCrossRefGoogle Scholar
  54. 54.
    Chevat C, Pena BM, Al MJ, et al. Healthcare resource utilisation and costs of treating NSAID-associated gastrointestinal toxicity: a multinational perspective. Pharmacoeconomics 2001; 19Suppl. 1: 17–32PubMedCrossRefGoogle Scholar
  55. 55.
    Sturkenboom MC, Romano F, Simon G, et al. The iatrogenic costs of NSAID therapy: a population study. Arthritis Rheum 2002; 47(2): 132–40PubMedCrossRefGoogle Scholar
  56. 56.
    Jonsson B, Haglund U. Economic burden of NSAID-induced gastropathy in Sweden. Scand J Gastroenterol 2001; 36 (7): 775–9CrossRefGoogle Scholar
  57. 57.
    Vargas E, Cabrera L, Moron A, et al. Assessment of hospital costs related to the diagnosis and treatment of upper gastrointestinal haemorrhages in patients consuming non steroid antiinflammatory drugs. Rev Esp Enferm Dig 2001; 93 (6): 390–7Google Scholar
  58. 58.
    Herings RM, Klungel OH. An epidemiological approach to assess the economic burden of NSAID-induced gastrointestinal events in The Netherlands. Pharmacoeconomics2001; 19(6): 655–65PubMedCrossRefGoogle Scholar
  59. 59.
    Rahme E, Pettit D, LeLorier J. Determinants and sequelae associated with acetaminophen utilization versus traditional NSAIDs in an elderly population. Arthritis Rheum 2002; 46(11): 3046–54PubMedCrossRefGoogle Scholar
  60. 60.
    Moore AR, Van Ganse E, Le Parc JM, et al. The Pain Study: paracetamol, aspirin, ibuprofen new tolerability: a large scale, randomised clinical trial comparing the tolerability of aspirin, ibuprofen and paracetamol for short term analgesia. Clin Drug Invest 1999; 18: 89–98CrossRefGoogle Scholar
  61. 61.
    Garcia Rodriguez LA, Hernandez-Diaz S. Risk of upper gastrointestinal complications among users of acetaminophen and nonsteroidal anti-inflammatory drugs. Epidemiology 2001; 12: 570–6PubMedCrossRefGoogle Scholar
  62. 62.
    Nyren O, Adami N-O, Gustavsson S. Social and economic effects of non-ulcer dyspepsia. Scand J Gastroenterol Suppl 1985; 109: 41–7PubMedCrossRefGoogle Scholar
  63. 63.
    Mietinnen OS. Theoretical epidemiology. Montreal (QC): John Wiley and Sons Inc., 1985Google Scholar
  64. 64.
    Coughlin SS, Benichou J, Weed DL. Attributable risk estimation in case controls studies. Epidemiol Rev 1994; 16: 51–64PubMedGoogle Scholar
  65. 65.
    Bentkover JD, Baker AM, Kaplan H. Nabumetone in elderly patients with osteoarthritis: economic benefits versus ibuprofen alone or ibuprofen plus misoprostol. Pharmacoeconomics 1994; 5(4): 335–42PubMedCrossRefGoogle Scholar
  66. 66.
    Brixner DI. A decision analysis model in the evaluation of NSAIDs in a managed care setting: a case study. Med Interface 1994; 7(11): 145–50PubMedGoogle Scholar
  67. 67.
    Goldstein JL, Larson LR, Yamashita BD. Prevention of nonsteroidal anti-inflammatory drug-induced gastropathy: clinical and economic implications of a single-tablet formulation of diclofenac/misoprostol. Am J Manag Care 1998; 4(5): 687–97PubMedGoogle Scholar
  68. 68.
    Liaropoulos L, Spinthouri M, Ignatiades T, et al. Economic evaluation of nimesulide versus diclofenac in the treatment of osteoarthritis in Greece. Pharmacoeconomics 1998; 14 (5): 575–88CrossRefGoogle Scholar
  69. 69.
    Liaropoulos L. Economic comparison of nimesulide and diclofenac, and the incidence of adverse events in the treatment of rheumatic disease in Greece. Rheumatology (Oxford) 1999; 38Suppl. 1: 39–46CrossRefGoogle Scholar
  70. 70.
    Peris F, Martinez E, Badia X, et al. Iatrogenic cost factors incorporating mild and moderate adverse events in the economic comparison of aceclofenac and other NSAIDs. Pharmacoeconomics 2001; 19(7): 779–90PubMedCrossRefGoogle Scholar
  71. 71.
    McCabe CJ, Akehurst RL, Kirsch J, et al. Choice of NSAID and management strategy in rheumatoid arthritis and osteoarthritis: the impact on costs and outcomes in the UK. Pharmacoeconomics 1998; 14(2): 191–9PubMedCrossRefGoogle Scholar
  72. 72.
    Pellissier JM, Straus WL, Watson DJ, et al. Economic evaluation of rofecoxib versus nonselective nonsteroidal anti-inflammatory drugs for the treatment of osteoarthritis. Clin Ther 2001; 23(7): 1061–79PubMedCrossRefGoogle Scholar
  73. 73.
    Marshall JK, Pellissier JM, Attard CL, et al. Incremental costeffectiveness analysis comparing rofecoxib with nonselective NSAIDs in osteoarthritis: Ontario Ministry of Health perspective. Pharmacoeconomics 2001; 19(10): 1039–49PubMedCrossRefGoogle Scholar
  74. 74.
    Zabinski RA, Burke TA, Johnson J, et al. An economic model for determining the costs and consequences of using various treatment alternatives for the management of arthritis in Canada. Pharmacoeconomics 2001; 19Suppl. 1: 49–58PubMedCrossRefGoogle Scholar
  75. 75.
    Chancellor JV, Hunsche E, de Cruz E, et al. Economic evaluation of celecoxib, a new cyclo-oxygenase 2 specific inhibitor, in Switzerland. Pharmacoeconomics 2001; 19Suppl. 1: 59–75PubMedCrossRefGoogle Scholar
  76. 76.
    Pettitt D, Goldstein JL, McGuire A, et al. 0verview of the arthritis cost consequence evaluation system (ACCES): a pharmacoeconomic model for celecoxib. Rheumatology (0xford) 2000; 39Suppl. 2: 33–42CrossRefGoogle Scholar
  77. 77.
    Haglund U, Svarvar P. The Swedish ACCES model: predicting the health economic impact of celecoxib in patients with osteoarthritis or rheumatoid arthritis. Rheumatology 2000; 39Suppl. 2: 51–6PubMedCrossRefGoogle Scholar
  78. 78.
    Svarvar P, Aly A. Use of the ACCES model to predict the health economic impact of celecoxib in patients with osteoarthritis or rheumatoid arthritis in Norway. Rheumatology (0xford) 2000; 39Suppl. 2: 43–50CrossRefGoogle Scholar
  79. 79.
    You JHS, Lee KKC, Chan TYK, et al. Arthritis treatment in Hong Kong: cost analysis of celecoxib versus conventional NSAIDS, with or without gastroprotective agents. Aliment Pharmacol Ther 2002; 16(12): 2089–96PubMedCrossRefGoogle Scholar
  80. 80.
    Régie de l’assurance maladie Québec. Liste de medicaments complete [online]. Available from URL: http://www.ramq.gouv.qc.ca/fr/professionels/listmed/lm_tdmf.shtml. [Accessed 2004 Sep 20]Google Scholar
  81. 81.
    Rothman KJ, Greenland S. Modern epidemiology. 2nd ed. Philadelphia (PA): Lippincott-Raven, 1998Google Scholar
  82. 82.
    Rahme E, Marentette MA, Kong SX, et al. Use of NSAIDs, C0X-2 inhibitors, and acetaminophen and associated coprescriptions of gastroprotective agents in an elderly population. Arthritis Care Res 2002; 47(6): 595–602CrossRefGoogle Scholar
  83. 83.
    Yeomans ND, Tulassay Z, Juhasz L, et al. A comparison of omeprazole with ranitidine for ulcers associated with nonsteroidal antiinflammatory drugs: acid suppression trial: ranitidine versus omeprazole for NSAID-associated ulcer treatment (ASTR0NAUT) study group. N Engl J Med 1998; 338(11): 719–26PubMedCrossRefGoogle Scholar
  84. 84.
    Burke TA, Zabinski RA, Pettitt D, et al. A framework for evaluating the clinical consequences of initial therapy with NSAIDs, NSAIDs plus gastroprotective agents, or celecoxib in the treatment of arthritis. Pharmacoeconomics 2001; 19(1): 33–47PubMedCrossRefGoogle Scholar

Copyright information

© Adis Data Information BV 2004

Authors and Affiliations

  • Elham Rahme
    • 1
  • Alan N. Barkun
    • 2
  • Viviane Adam
    • 3
  • Marc Bardou
    • 4
  1. 1.Department of Medicine, McGill University and Research InstituteMcGill University Health CenterMontrealCanada
  2. 2.Department of Medicine, Division of GastroenterologyMcGill University and the McGill University Health CenterMontrealCanada
  3. 3.Research InstituteMcGill University Health CenterMontrealCanada
  4. 4.Department of Medicine, Clinical Pharmacology UnitFaculty of MedicineDijonFrance

Personalised recommendations