Safest 2015 SUVs and Minivans with Three Rows of Seating for Families

8 Comments

1. SafeDad - CPS Darren January 16, 2016

   Here are our 2016 Winners of Safest Family Minivans/SUVs: https://carseatblog.com/36788/safest-2016-family-suvs-and-minivans-with-three-rows-of-seating/
2. CPSDarren February 3, 2015

   Hi Andrew,

   I don’t believe the IIHS injury/fatality data provides any comparable evidence between the current generation Acura MDX and LC/LX570. Even the just-released driver fatality data proves that they cannot account for all demographic data that isn’t related to safety. This is not only in discussed the text, but like in past studies, is shown again by significantly varying results of essentially identical vehicles. The relatively heavy Mercury Grand Marquis is about twice the average rate for all vehicles, while the identical Ford Crown Vic has a rate nearly zero. Tall, heavy vehicles like the Ford Expedition 2WD and Chevrolet Suburban 2WD had above average driver fatalities, while the same models with 4WD had lower rates. Proof that 4WD is safer than 2WD? Perhaps, except that the Chevy Silverado 4WD is one of the worst performers in the study, while the 2WD equivalent is well below average. Meanwhile, lighter, shorter vehicles like the Subaru Legacy sedan and Audi A4 had zero fatalities, apparently proving them just as safe as models like the Volvo XC90 if you believe these results.

   I’d like to think my own 2011 Highlander Hybrid is among the safest vehicles for driver fatalities given its driver death rate of zero, and perhaps it is, but it is impossible to separate the type of buyer/driver from the actual safety of the vehicle. These insurance claims data for injury claims and driver fatality risk are simply not good tools for comparing individual vehicles as you suggest. The trends for mini, small and sports cars are well documented in terms of higher death rates, no doubt at all about that. For weight, however, the advantage begins to level off at a point around the midsize vehicle classes if you believe the IIHS study. For example, all classes of SUVs, small to large, were essentially average or below average. This contradicts your claim of the LX570 being safer than an Acura MDX based only because the LX is a heavier vehicle. The study itself simply shows too many discrepancies to make such a claim. For example, small 4WD SUVs had a lower risk than the very large 2WD models, while large 2WD SUVs had a lower risk than small or midsize 4WD models. Very large pickups, among the heaviest and tallest vehicles, were among the classes of vehicles that had higher than average driver death rates. In fact, all classes of pickups had essentially the same driver risk as midsize and large sedans, showing that being heavier and/or taller is not always safer. There are many other factors involved in overall safety.

   I really wish it could be proven that vehicles were identical in all markets, but it is simply impossible. Even manufacturers that show trends of having better crash test performers have vehicles that perform differently in testing. Ultimately, if you really want to compare apples to apples in the USA in terms of crash protection and crash avoidance, you do have to rely upon IIHS and NHTSA crash testing results for USA spec vehicles. You can guess that untested models will do well or do poorly, based on trends for manufacturers or tests in other countries, but it is only a guess.

   As for the Acura MDX that rated tops in this survey, nimble handling and being child seat friendly were not among the reasons it scored well, though personally I think they are also important things for some to consider. The full complement of available crash avoidance safety features (mostly lacking in the LX) and a perfect set of IIHS and NHTSA crash test results are the main factors in the survey. That’s what puts the new MDX in a different league than the LX or even my own Highlander Hybrid, despite the driver death rates observed for the last generation models.

   And that is my point. When creating a list of the safest vehicles, speculation makes for good debates. But crash test results and advanced safety features are directly comparable and far more objective. Yes, some very safe vehicles are excluded because they lack advanced safety features and/or crash test results. There is a very easy way for manufacturers of those vehicles to remedy these omissions, however!
   • Geoff October 22, 2015

     Darren,
     Thanks for this post. I appreciate Andrew’s points, but you hit the nail on the head. If Toyota wants me (and thousands of other safety-conscious parents in the USA) to buy an $80k Land Cruiser, PROVE to me it’s crash-worthy. Put your money where your mouth is.
3. Andrew January 22, 2015

   Hi Darren,

   Thank you for your thought out response.

   I agree that your method is objective and repeatable. My concern is that by overvaluing American crash test results, people could be misled into thinking that a small 5 star rated vehicle is safer than a large 4 star rated vehicle.

   I agree with you that the Mercedes GL is one of the safest vehicles you could buy. I didn’t mention it because there is no crash test data on it that I’m aware of, but the Mercedes pedigree for crash testing is as strong as they come. I read an article last year about the strained partnership between Mercedes and Nissan/Renault after French engineers mistakenly placed a seatbelt in front of a side airbag on a Mercedes badged Renault van. It resulted in the worst star rating (only 3 stars) that Mercedes had received in 15 years. The GL may have no crash test results, but it comes from a manufacturer that is renowned for safety design. If you look through the last 3 published IIHS studies on driver death rate, the manufacturers that did outstanding across a range of studies (1999-2008) were Mercedes (E-class, M-class), Toyota/Lexus (4Runner, RX, Sienna) and Honda (Pilot, Odyssey). Other than the E-class, these are all SUVs or minivans.

   I also agree that death rates and injury rates cannot be completely statistically adjusted for all of the different factors that come into play. Even the same model has some variation from one year to the next. And, as you pointed out, there are gaps in the data that we have (especially for new models).

   However, the data that we do have shows that the type and brand of vehicle purchased is extremely important. For example, full size Japanese pickup trucks only appeal to a certain demographic. In 2008, Americans who purchased a new Toyota Tundra (a Top Safety Pick) had an adjusted driver death rate of 23 deaths per million, making Tundra drivers among the least likely to die in the nation. Americans who purchased a new Nissan Titan had a death rate of 111-126, putting them among the least safe drivers in America. A used Titan may look just as big and safe as a used Tundra, but the safety difference is dramatic.

   By studying the injury rates and driver death rates (even those with fairly broad 95% confidence intervals), and then incorporated worldwide crash test results to help explain the trends in the data, it is possible to derive a good overall understanding of the relative safety of different vehicles.

   As I pointed out above, the trends show that bigger, heavier vehicles have lower injury rates, and that is what is critical from a child safety perspective. There are multiple factors which affect a child’s injury likelihood; two affected by the vehicle design are cabin intrusion and acceleration. Cabin intrusion is related not only to side impact scores (which is why I would not buy a full size van) but also by the height of a vehicle. Car seats in SUVs tend to perch above the crush zone of other vehicles, keeping children safer. Acceleration forces will vary based on the mass of the vehicle that the car seat is in (f=ma). HLDI injury rates show a general downward curve of about 20% fewer injury claims per 1000 pounds of curb weight.

   As an example of the importance of vehicle size and weight, the NHTSA commissioned a study to measure concussion likelihood in side impacts for children in car seats. The difference between side impact HIC scores (concussion likelihood) for the child varied more based on whether the seat was installed in an SUV or a small car (Nissan Versa/Sentra) than the difference between the seats that did well in the proposed side impact crashes (e.g. Britax Advocate) and those that flunked them. (See “Development of a Lateral Test Procedure for Child Restraints,” slide 14 http://www.nhtsa.gov/Research/Child+Seat+Research ) Of course, the good news for future babies is that the NHTSA took the Nissan Sentra/Versa side intrusion profiles and used it as the proposed baseline test for all carseats in the NPRM on side impact testing. So future car seats in SUVs and minivans will be safer because they’re assuming worst case scenario (installation in a small Nissan).

   In regards to manufacturers cutting corners and having poorer performance from one country to the next, I would argue that global NCAP results show a big difference between the manufacturers that truly value safety and those that only pay it lip service. In the Latin NCAP scores, the best manufacturers were Honda, Toyota, and Ford (with the exception of the Skoda built Ford Ka). Chevrolet had some poor results, but the cars that aced the IIHS moderate offset test also aced the Brazilian moderate offset test, showing that Chevy is at least being honest. The only manufacturer that had a significantly worse results between the European moderate offset and the identical Brazilian moderate offset was Renault/Nissan, which (if you’re still reading) is not a company I have a high regard for. They are the only manufacturer that I know of that has been caught substituting cheaper steel into the safety cages of vehicles they sell in other countries.

   Tests around the world have shown that certain manufacturers truly value the safety of their occupants. Honda, Toyota, Ford, Mercedes and Volvo all create vehicles that have consistently excelled. The vehicles that are created on the same assembly line (e.g. the Land Cruiser made exclusively in Japan or the XC-90 made exclusively in Sweden) are vehicles that I would trust in any country.

   As far as getting an older one without stability control, no I wouldn’t recommend that either. But the last year you could get a Land Cruiser without stability control was 1999, and that was a long time ago. A 2003 LC or XC-90 (or an older GL-class) in rust free condition is (in my opinion) safer than the vast majority of new cars available today. And in terms of taller vehicles being rollover prone, that’s an outdated perspective that has been virtually eliminated with the combination of stability control, stronger roof pillars, and side/curtain airbags. In fact, a few years ago the IIHS tried to find a fatal rollover accident in a vehicle with those three features and couldn’t do it.

   Here’s my final point, and then I’ll conclude. Yes, there’s a lot of confounding data in the results that we have, and no, you cannot look at death rates or injury rates in absolute terms. However, even a crash test usually has a sample size of only one, and that’s why I think it’s better to look at as much data as possible. For example, the second generation Dodge Caravan had a 4 star safety rating from the NHTSA in a frontal crash. The Europeans tested it in a moderate offset and said it was only a 2 or 3 star rated vehicle (depending on steering wheel location). The IIHS tested it in a moderate offset and rated it “Poor”, then retested it and got a better result but discovered a horrible fuel leak. The third test showed that the leak had been fixed and earned an “Acceptable” rating. The dummy’s head hit the steering wheel through the airbag in the first and second tests, indicating that concussions were possible. The left femur could have broken on the first and third test, and the right leg would have been injured in all three tests. What’s my point? That each of the 5 published moderate offset tests show slightly different results. Conversely, the IIHS measured driver death and injury rates multiple times and found that Grand Caravan / T&C owners were two to four times more likely to die than Honda Odyssey owners until the 2008 redesign, and have had higher injury rates (which is still true even now).

   Are crash tests important? Absolutely. But they only tell part of the story. A crash test generally shows how crashworthy the design is, and how safe it is compared to other vehicles of the same weight or smaller. A heavier vehicle with stability control is generally safer than a lighter vehicle, and taller vehicles tend to have fewer occupant injuries than lower vehicles. Taking all of that together, I would argue that the 2008+ Land Cruiser/LX570 is safer than an Acura MDX, even though the Acura is more nimble and has more LATCH and tether points. But that’s just my opinion, and I too have certainly been wrong before :-).
4. Pingback: Safest Family Sedans for 2015: Part III January 5, 2015
5. Pingback: Safest 2015 5-passenger SUVs for Families: Part II December 29, 2014
6. CPSDarren December 21, 2014

   Hi Annie, the Mazda 5 is a very nice family hauler; we have a review of it here at CarseatBlog. In addition to the Mazda5, the Dodge Grand Caravan American Value Package is another bargain family hauler. For our winners, we did emphasize very good crash test scores and crash avoidance features in our list and unfortunately these sometimes come at a price. For example, neither the Mazda5 or Grand Caravan has a frontal crash prevention system available, though the Dodge did turn in reasonably good results in the NHTSA crash tests and also in the IIHS testing except for the small overlap test. The Mazda5 is untested by the NHTSA and did not perform well in the IIHS testing.
7. Andrew Walton December 21, 2014

   Hi Darren,

   This is a great list, and it’s hard to argue with the Acura MDX. However, I feel like requiring NHTSA crash test data is depriving your readers of some great selections. Two great choices that are 5 star vehicles in Europe and Australia but don’t sell enough for the NHTSA to test them are the Volvo XC-90 and the Toyota Land Cruiser.

   The vehicles you included are all extremely safe vehicles. However, vehicle safety can be measured in three ways: crash tests (NHTSA, IIHS, Euro NCAP, Australian NCAP), IIHS driver death rates, and HLDI injury rates. Frontal crash tests show how well a safe a vehicle would be in a collision with the identical vehicle. They are generally good for seeing how well a vehicle would do against a similar or smaller vehicle. Driver death rates have shown for the last several years that the safest vehicles (least likely for the driver to die in) are midsize to large SUVs and minivans. Injury rates show that the safest (least likely to be injured in) vehicles are heavy SUVs and pickup trucks.

   The IIHS this summer released their list of top used vehicles for teens, noting among other things that “bigger, heavier vehicles are safer” and stability control (which “reduces risk on a level comparable to safety belts”) is a must. Their top value pick for safety was the 2005+ Volvo XC-90.

   The new XC-90 will be even safer. In fact, pound for pound, I doubt there is a safer vehicle on the road. The last Volvo SUV to be released, the XC-60, had the best crash test scores Euro NCAP had ever seen.

   When the IIHS implemented the new small offset test it required major modifications and or redesigns from the safest manufacturers. Honda, Mercedes and Jeep all augmented their designs before testing. Toyota didn’t until several models completely flunked, and they went back to the drawing board. Volvo, on the other hand, took the XC-90, a ten year old design, and aced the test. Why? Because Volvos (like Honda and Mercedes) are built for safety, not just passing the test. When Ford bought Volvo in 1999, Ford vehicles were some of the worst in America, with driver death rates comparable to Chevy, (though not as bad as Nissan or Mitsubishi). Less than ten years after buying Volvo and adopting much of their safety technology, Ford drivers became as safe as drivers of Honda, Toyota, Mercedes or Jeep. Simply redesigning the F-150 so that the engine compartment (instead of the passenger compartment) would crush in a moderate offset cut the driver death rate in half.

   However, as the IIHS points out, size and weight do matter. Frame height and preventing override are the reason why SUVs are seven times safer than cars in head on collisions (4x safer if the car had a better NHTSA rating, 10x safer if the SUV had the better rating).

   http://youtu.be/ExQUGk12S8U

   http://www.buffalo.edu/news/releases/2013/05/026.html

   http://www.iihs.org/iihs/topics/driver-death-rates

   http://www.iihs.org/iihs/topics/insurance-loss-information

   Injury rates show a decrease of injury frequency by about 20% per 1000 pounds. So after statistically adjusting for age, gender, marital status, location, date, etc., drivers of the Toyota Land Cruiser/LX had an injury rate of roughly 5.5 per 1000 registered vehicles per year. The XC-90 had a rate of 8.3, the MDX 8.8, the Odyssey and Sienna were around 12, the Mazda 5 around 14, and the Nissan Versa (for comparison purposes) 25.2.

   What’s my point? My point is that simply because a vehicle does well in American crash tests doesn’t mean that it is the “safest”. For children, the most important aspects of crash testing are deceleration forces and cabin intrusion, neither of which a crash test can predict if you are in a collision with a larger vehicle.

   If money was no object and I wanted the safest 3 row SUV I could buy, I would probably invest in a 2008+ Toyota Land Cruiser or Lexus LX570. It is the heaviest vehicle to have a 5 star rating from a country that uses the moderate offset test, has a relatively high (and very stiff) frame, and comes from a manufacturer that has not made a poor (safety wise) SUV since the introduction of electronic stability control.

   If you’re not convinced that size trumps crash test data, than I would get the 2016 XC-90.
   • CPSDarren January 6, 2015

     Hi Andrew, thank you for the great comments! I tried to keep the methods as simple as possible, to make them as objective as possible. As I said in the article, there are plenty of very safe vehicles and most midsize crossover SUVs and minivans provide excellent protection for occupants. But if you’re looking for a short list of the very safest models based on simple criteria from the IIHS and NHTSA, I list only those with proven results to be the very safest overall.

     Of course, there is always potential for debate. And that’s a good thing, because there are many excellent models that didn’t make the cut. I’d add the Mercedes GL-Class as one of my personal favorites that didn’t have a full set of crash tests. But keep in mind that the manufacturers can and do arrange for priority testing of vehicles with the NHTSA and IIHS at their own cost, especially for new models and those that may not otherwise be tested. That some companies don’t do this is likely because they are cutting costs or concerned about the possible results. In that I have no way to determine which, I chose to stick with models that earned top ratings in all available tests in the USA.

     Crash testing results from Europe, Australia and Japan are all interesting and valid in those countries, but not necessarily relevant in the USA. That is because there is no guarantee that models sold there are exactly identical to those sold in other markets. Simple cost cutting measures can remove a reinforcement, an airbag, a seatbelt pre-tensioner or other feature critical to safety and/or crash testing results in one or more countries. Manufacturers may also de-content to meet only minimum federal standards in a particular country, while some make changes to safety features just for marketing purposes. Maybe a model like the XC90 is exactly the same in every way across all markets, but I have no way to determine this. To compare apples to apples for this article, I did not consider overseas testing results.

     IIHS/HLDI death and injury rates are flawed for direct comparisons on new models. They simply don’t exist for new 2014-2015 models, only on older models that may vary substantially in features or even design. Even for models that have not changed, like the XC90, these rates have a wide margin of error. Though they try to compensate for driver demographics, you can see this margin by comparing identical corporate twin vehicles over the years that often have varying results. While there is an element of crash safety in these statistics, it is impossible to use them for direct comparison. What they really measure is claims data, that is only indirectly tied to crashworthiness and crash avoidance. That’s good for showing trends, like the safety of midsize SUVs compared to compact cars, but not for directly comparing midsize SUVs with each other, like the ones in this article.

     As you mention, weight is definitely a factor for frontal crashes with another vehicle, tapering off for the heaviest vehicles, but weight is not as much of a factor in side impacts. On the flip side, tall, heavy truck-based vehicles with a stiffer chassis often do worse in single vehicle crashes, such as rollovers that have a high incidence of fatalities. Seated passenger head height level is important for side impacts, especially the IIHS test that uses a shorter dummy. Even so, most models mentioned in this article are pretty similar in terms of weight and height, so it’s not really a factor like it would be if I had included sub-compact sedans, for example.

     I have no doubt the XC90 is a very safe vehicle. It was designed over a decade ago by Volvo engineers prior to and during the transition to Ford, and remains essentially the same today. Is it as safe or safer than models that made this list? Perhaps, but we’d only know for sure if Volvo had requested that the NHTSA test it at any time since their crash testing was updated for 2011! It is worthwhile to note that the company has changed hands twice, and now is owned by a Chinese interest with yet another management change. Even without those changes, there is no guarantee that the upcoming 2016 model will perform as well in terms of safety, but hopefully Volvo continues to put an emphasis on crashworthiness and crash avoidance features. The XC60 you mention is another extremely safe vehicle, and was included in the next part of this series for 5-passenger SUVs.

     As for the LX570 and Land Cruiser, well, I wouldn’t want my family in one daily compared to other choices available. Why? https://carseatblog.com/27619/bigger-is-not-always-better-lexus-lx570-video-review/ For overall safety, I’d take the new MDX w/Advance package, or even a Mercedes GL or Audi Q7, over the LX any day, for that matter. Aside from crash tests and crash avoidance features, the seating flexibility for carseats including top tether anchors is much better in the MDX. Like the XC90, the Land Cruiser and LX570 completely lack important top tether anchors in the third row. Plus, routine and emergency handling in the LX570 is dismal compared to the MDX. I wouldn’t even consider a purchase of an older Land Cruiser that lacked stability control and side curtain airbags with rollover sensors. But that’s just my opinion! I’ve been wrong before:-)
8. Annie December 19, 2014

   I just bought a Mazda 5 and I am quite happy. 6 passsengers with sliding doors I feel I get commodity and good mileage.

   I had looked at differents models and the ones you mention are too pricey for me!