tl;dr version for those uninterested in the longer version: Provisionally yes, more selectivity will usually result in fewer side effects. It can also result in a greater chance of some unusual other side effects, like sleepwalking and other parasomnias, and may not be any less addictive than the benzodiazepines (in fact, it can be argued to be on par or slightly more addictive).
The GABA type A (heretofore referred to as GABA-A) receptor complex is made up of 5 subunits. There are a huge number of possible subunit combinations since there are six different alpha subunits, at least three beta subunits, at least 3 gamma subunits, and a delta, epsilon, theta, pi, and 3 rho subunits. Happily, the vast majority of subunits tend to combine in roughly the same groupings in various parts of the brain. The most common combination found almost everywhere in the brain is what was once known as the BZ1 (benzodiazepine 1) receptor grouping, which is two alpha-1 units, two beta-(2/3) units, and one gamma-2 unit, in the pattern of an alpha-beta-alpha-beta-gamma ring, with the first alpha unit in said pattern interacting with the gamma unit. Benzodiazepines and the Z-drugs (zolpidem, zaleplon, and (es-)zopiclone) fit into a specific pocket in between the alpha-1 and gamma-2 subunit and favorably improve how well the GABA molecule can bind to and open the GABA-A ion pore.
Zolpidem binds about 10-20x more strongly if the alpha unit is alpha-1 than if it is alpha-2 or alpha-3 (neither benzodiazepines nor z-drugs bind to alpha-4 or alpha-6 containing GABA-A receptors. Ethanol, on the other hand, does bind to extrasynaptic alpha-4-beta(1-3)-delta GABA-A receptors) and it binds around 100x (if I’m remembering correctly) as strongly to alpha-1 as compared to alpha-5 containing GABA-A receptors. The benzodiazepines, in contrast, bind equally well (so far as we currently know) to all known benzodiazepine sensitive combinations (anything with a gamma-2 and alpha-1, 2, 3, and 5 subunit). Thus, zolpidem, at the standard FDA approved doses, is largely selective for GABA-A receptors with alpha-1 subunits, where clonazepam, temazepam, and alprazolam, to name a few benzos, would not be.
This selectivity would then manifest with fewer side effects, in theory, than the traditional benzodiazepines, since it isn’t augmenting GABA transmission at as many receptor subtypes. In practice, this seems to hold true, but does introduce a potentially greater risk of the more newsworthy unusual side effects (sleep-walking/driving/eating/sex). One of the benefits of using zolpidem at FDA approved doses, for example, is that it shouldn’t actually induce much, if any, dependence, since the dependence producing characteristics of benzos are speculated to be linked with alpha-2 and alpha-3 subunit containing GABA-A due to where those type of receptors tend to be located in the brain (amygdala, hypothalamus, striatum, and locus ceruleus, to name a few regions for those interested) and that those types tend to be post-synaptic (the neuron receiving a signal), whereas alpha-1 containing units can be pre-synaptic auto-receptors as well. Of course, no drug is perfect, so while it shouldn’t (at the FDA labelled doses) induce dependence (and the majority of studies back this up), it can still lead to addiction.
Confused about my use of the two underlined terms? Don’t feel bad, the two are often mistakenly conflated, including by many healthcare practitioners, in my experience. Dependence is a state where your body has gotten used to the presence of the drug and has adapted to this presence in myriad ways, resulting in a withdrawal syndrome of some kind if the drug is removed (or it’s effects blocked by an antagonist–in the case of the benzos and z-drugs, that would be an iv medication known as flumazenil). This is often linked to drug tolerance, which is where an amount of drug which once produced an effect is no longer enough to produce that same effect resulting in the need for a larger amount. Note that while they are often linked, tolerance is not necessary for dependence to develop and vice-versa. Addiction is a complex bio-psycho-social disorder characterized by compulsive seeking (craving) or taking, despite the harm it does to the person (and often their circle of friends and family). Addiction is, in essence, a set of maladaptive memories which are unusually powerful in provoking certain habitual behaviors. It can also be thought of as a reward deficit disorder. Addiction is often reinforced by dependence (more specifically, the symptoms of drug withdrawal act as negative reinforcement to keep taking the drug) and many addicts are very tolerant to their drug(s) of choice at the peak of use, but addiction can (re-)occur without either dependence or tolerance being present at all, and one can be extremely tolerant and exhibit dependence to a drug without being addicted (this is very often where I see other healthcare practitioners get it wrong).
Anyways, back to zolpidem. Zolpidem doesn’t do much at normal doses at alpha-2 and alpha-3 subunit containing GABA-A, which are speculated to be the primary receptor forms mediating anxiolytic (anxiety terminating), anti-convulsive (seizure terminating), and autonomic fight-or-flight blunting effects. Since it doesn’t augment these functions, you don’t find a rebound effect (ie heightened anxiety, greater chance of seizures, over-active autonomic nervous system drive) from those on withdrawal of the drug like you can with the benzos. Unfortunately, it still can be addictive because the brain circuitry involved in addiction is different than the brain circuitry involved with dependence. Ordinarily, in a part of the brain known as the ventral tegmental area, GABAergic (GABA releasing) neurons tonically suppress the firing of Dopaminergic neurons which are part of the brain’s reward circuitry, by acting post-synaptically on alpha-2 or alpha-3 subunit containing receptors. This tonic suppression can be countered in a number of ways, one of which is by the presence of too much GABA triggering pre-synaptic alpha-1 containing auto-receptors which signal to the GABA releasing neuron to stop/release less. Unfortunately, since zolpidem augments GABA to be more effective at the receptor, and since it doesn’t bind all that well to the post-synaptic GABA receptors for reasons already mentioned, zolpidem enhances the suppression of the GABA releasing neuron without also somewhat blunting/mitigating that effect by also augmenting post-synaptic GABA suppression like a benzodiazepine would, resulting in the “reward” dopaminergic neuron firing more frequently. While this is, by itself, not enough to cause addiction, the current thinking (or at least, the thinking I’ve read up to this point) is that the over-active firing of these dopaminergic neurons is a very key part of the overall picture involving the neurobiology of the development of addiction.
That’s probably more of an answer than you expected or wanted, but this particular area fascinates me and I couldn’t help myself. 